Cybernetics of Living Matter: Nature, Man, Information

Citation preview

Cybernetics

of Living ^ Matter: E™.™, Editor I.M. MAKAROV

!

/

\

^

1 I

Cybernetics of Living Matter

Kn6epHeTMKa >KMBOrO noA peAaHuweM HJi.-Kopp. AH C C C P

H. M. ManapoBa

CocTaBHteJib — B. fl,. flenejiHC H3Aaie;ibCTB0 «Mwp» MocHea

Cybernetics

of Living Matter: Nature,

Man, Information

Editor

I. M. MAKAROV, Corresponding Member of the USSR Academy of Sciences Compiled by V. D. Pekells Translated from the Russian by V. I. Kisin

MIR PUBLISHERS MOSCOW

First published 1987

Ha

O M AU UCKO M A3blK e

Printed in the Union of Soviet Socialist Republics

© Mir Publishers, 1987

Contents

Preface. Yu. A. Ovchinnikov 7 Note from the Compiler. V. D. Pekelis 10 I. Biology and Information 23 Biology Today 23 Basic Tendencies in Physico-Chemical Biology. Yu. A. Ovchin­ nikov 23 Genetics, Evolution, and Theoretical Biology. N. V. Timofeev-Resovsky 33 Transition to Constructing Living Systems. A. A. Baev 41 Autowaves: An Interdisciplinary Finding. G. R. Ivanitskyt V. I. Krinsky, and 0. A. Mornev 52 Cybernetics' Standpoint 75 Cybernetics Approach to Theoretical Biology. A. A. Lyapunov 75 Information Tneory and Evolution. M. V. Volkenshtein 83 Control Sciences and the Harvest. Yu. M. Svirezhev 94 II. The Complexity of Living Systems 105 Integrity of Life 105 On Systematic and Integral Nature of Man. V. G. Afanasyev 105 The Contribution of Psychology to Systems Research of Man. B. F. Lomov 115 Brain and Intelligence 127 Natural Intelligence versus Artificial Intelligence: The Phi­ losophical View. P . K. Anokhin 127 On Reliability of the Brain. A . B . Kogan 142

6

Contents

Novel Aspects 154 Diurnal Rhythms and Adaptation. V. N. Reushkin 154 Searching Activity, Sleep, and Stability of the Organism. V. S. Rotenberg 183 On Man's “Third State”. V. /. Klimova 195 III. Difficulties on the Road to Truth 203 Science and New Information 203 Road to Truth (on the scientific method of cognitionf. A. B . Migdal and E. V. Netesova 203 The Dynamics of New Truths in Biological Sciences. S. E. Shnol 217 On New Knowledge in Biological Studies. B. V. Biryukov 229 Criteria of Existence and Conflicting Situations in Science. D. I. Dubrovsky 237 New Horizons in Cognition 244 The Physical Fields of Biological Objects. Yu. V. Gulyaev and E. E. Godik 244 Man’s Magnetic Fields. V. L . Vvedensky and V. /. Ozhogin 25 Radio Freguency Emission of Human Body and Medical Diag­ nostics. V. S. Troitsky 266 Several Problems in Psychology 276 The Psychology of Cognition and Cybernetics. B. M. Velichkovsky 276 Subconsciousness and Superconsciousness. P. V. Simonov 292 The Principle of Active Operator in Engineering Psychology. B . F. Lomov 307 The Organism and Age 325 Ageing and Old Age. V. /. Klimova 325 Extension of Human Life: The Biological Dimension and Expe­ rimentation. V. V. Frolkis and Kh. K. Muradyan 336 Overhaul of Man. V. D. Pekelis 352 About The Authors 361

Preface

The breathtaking discoveries of today’s biology not only revolutionize our view of the living matter but also make a profound impact on medicine, agriculture, and a number of manufacturing industries. Had the ideas and methods of cybernetics not been taken over by the biological theory and its applications, these discoveries would, however, be in­ conceivable. Control science, theory of large-scale systems, informa­ tion theory, studies of data transmission systems and com­ munication channels in the living matter coupled with the ideas and techniques coming from chemistry, physics, and mathematics shape the biological science of today. All of its numerous '‘narrow”, “specialized”, or “traditional” fields have experienced a profound impact of cybernetics; its heuristic fruitfulness, now indispensable for biology, is obvious in all of them. Extensive utilization of experimental methods, simula­ tion studies, and systems analysis promoted biology to the rank of exact science. Some people have probably forgotten that biology used to be a descriptive discipline. The successes scored by this science are impressive and widely acclaimed. For illustration, I will take up my own line of research, physico-chemical biology. Soviet scientists have significantly contributed to un­ raveling the basic mechanisms of storage and expression of genetic information, to discovering the laws of regulation

8

Yu. A. Ovchinnikov

and energy supply in a living cell, and to studying the struc­ ture and chemical synthesis of substances. Genetic engineering which came into existence about ten years ago has been fruitfully developing in the USSR. This new science makes it possible to change in a purposeful way the machinery of inheritance, to “design the living matter”. Biotechnological processes are utilized in the manufacture of medical drugs, foodstuffs, and fodder. A major task for our science today is to expand and deepen basic research and to make its findings work in practical fields. + The humanitarian nature of today’s biology should by no means be overlooked. It is essential for science as an entity and especially for genetics and psychophysiology and for research on the structure and functioning of the brain. Narrowly specialized researchers who, by force of tradi­ tion, explore the unknown by advancing along their cherished “rut”, usually find themselves in a cul-de-sac. They may ob­ viate this predicament when they study something less com­ plicated but in trying to solve the mysteries of Man the scientist should deliver the attack in the most interdisciplin­ ary fashion. For this reason the “cybernetics of the living matter” cannot be treated apart from various aspects of its application to studies of man. Every scientist is aware of the pressing need in compre­ hensive studies of man. Even if a far cry from approaching the answers, every step forward must be thoroughly report­ ed. The vital importance of every research project in this field, particularly those of Soviet scientists, cannot be over­ emphasized. Man will remain a mystery unless certain basic problems are resolved, problems of such magnitude that scientists of the whole world pull their efforts together to attack them. The reader will recall that human thought developed most

Preface

9

successfully whenever the new knowledge was jointly shared and comprehended. This collection of papers addresses a wide range of read­ ers. The title of the book is hardly surprising. The import­ ance of cybernetics approaches to biology has been widely appreciated in the last decades and found overwhelmingly fruitful. I believe that this book reporting the cyberneticsaided findings of Soviet researchers in diverse biological fields will give the reader a wider view of this science. Academician Yu. A. Ovchinnikov, Vice-President of the USSR Academy of Sciences

Note from the Compiler

This collection of papers, as the reader will see from its title, “Cybernetics of Living Matter: Nature, Man, Infor­ mation”, dwells upon the fields of biological cybernetics and data processing in physiological systems. This choice is explained in the Preface and in the opening article written by a leading Soviet authority in biology AcSdemician Yu. A. Ovchinnikov, Vice-President of the USSR Academy of Sciences: “The scope of interests and problems in today’s biology is extremely wide: from elementary processes in a living cell to the development of the entire organism, to its interaction with other organisms and the environment in the ecological system. Today’s biology is rapidly evolv­ ing branch of science, rich in exciting problems and pros­ pects, commanding an army of enthusiasts, and armed with the most advanced techniques and equipment. Biology holds key positions in solving the global problems mankind con­ fronts, be it the battle against fatal disease, the food crisis, or the pollution of the environment”. It was long ago that cybernetics “took note” of biology while mathematical exploration of biology has a still longer history. Enhanced by the intersection of various fields of knowledge, penetration of mathematics into biology was especially vigorous during the last two or three decades. For a long time biology has been influenced by new ideas of physics and chemistry. The reader will recall the once sensational “What is Life? The Physical Aspect of the Liv-

Note from the Compiler

11

ing Cell” by E. Schrodinger, “The Kinetic Fundamentals of Molecular Biology” by a group of scientists, and “Molecules and Life” by M. V. Volkenshtein, Corresponding Member of the USSR Academy of Sciences, one of the contributors to this collection, to understand that in treating various issues in biophysics, biochemistry, and molecular biology prob­ lems arise which are akin to those in theoretical physics and chemistry. Application of physical and chemical find­ ings to biological research naturally results in a mathemat­ ical way of thinking and in the use of techniques borrowed from informatics and cybernetics. Today biological stud­ ies are unthinkable without mathematical tools, control theory, and information theoretic concepts. It is a remarkable fact that biology has been not only a field of application for cybernetics where the potential of new theories was tested but also an aid in making new discoveries. Biology has proved very helpful in artificial intelligence, pattern recognition, robotics, and other fields of research. The complexity of such phenomena as the func­ tioning of the brain, interactions in biological communities, adaptability, reproducibility, survival of living organisms, and their high reliability called for more sophisticated mathematical models and new control procedures. Thus search for mathematical modeling of reproduction processes culminated in the theory of self-reproducing automata. A complete list of cybernetic applications to biology would he impressive by its mere size. We will name just a few: studies of principles underlying the control of physiological processes in an organism as an entity; analysis of control and regulation mechanisms in physiological systems such as hlood circulation, respiration, and exchange of matter and energy; systematic representation and mathematical mod(ding of evolutionary processes; studies of control and data processing mechanisms in embryogenesis and growth

12

V. D. Pekelis

of the organism; studies of reliability in living nature; various levels of studying the oscillatory process as a major principle embodied in the organization of biological systems; analysis of the behaviour of living beings in various environ­ ments; unraveling the mechanisms of the brain functioning; and development and manufacture of biotechnical assemblies which would act as live organs. Cybernetically, the most challenging lines of research are the control mechanisms and data processing in living beings and possible applications of information-theoretic me­ thods to studying the functioning of human sensory organs, nervous system, and behaviour. Of course, a small book cannot cover all these aspects. The reader will find in it several enlightening articles on controversial issues. Some of the articleg*provide overviews of their fields while others describe specific research projects. In addition to a survey of general trends in physico­ chemical biology made by Academician Yu. A. Ovchinni­ kov, this collection includes an article reporting an applied study, “Autowaves: an interdisciplinary finding”. In this article a merger of physico-chemical and informatics ap­ proaches to biological phenomena is shown to give rise to a new line of research which results in the discovery of fascinating world. The new classical research of A. A. Lyapunov, Correspond­ ing Member of the USSR Academy of Sciences, on a cybernet­ ic approach to theoretical biology is followed up in the col­ lection by Yu. M. Svirezhev’s article “Control sciences and the harvest” in which he tried to demonstrate the use of “non-conventional” tools in agriculture which is viewed as an inhomogeneous complex system consisting of energetic, economic, ecological, and informational components. He proposes a technique for determining the conditions under which the system will operate in an optimal way.

Note from the Compiler

13

The much-discussed and ever-relevant issue, the evolu­ tionary theory, is viewed from different standpoints by N. V. Timofeev-Resovsky in his “Genetics, evolution, and theoretical biology” and M. V. Volkenshtein in his “Informa­ tion theory and evolution”. The former shows that a bet­ ter understanding of the evolution largely depends on the development of a general theory of biology and the latter concentrates on the value of information in biological struc­ tures in the course of evolution. Both authors analyze a variety of complex scientific data on this very important issue. Academician B. S. Sokolov argues in his review “A half century of thinking in biology” of A. A. Lyubishchev’s book “On ways to systematize the evolution of organisms” that the discouraging fact in some biological fields such as the evolutionary theory is excess, rather than shortage, of various ideas. Indeed, biology abounds in various des­ criptions, observations, experimental results, and models. This flow of information cannot be handled by the avail­ able processing tools. The overriding need of biology is now to evaluate the “product” supplied to the information “market”. For this reason, according to B. S. Sokolov, cri­ ticism has a major role to play. This criticism “manifests it­ self in different ways and on different levels, from the pro­ cedures of specific research projects to philosophic interpre­ tation of scientific activity as a whole”. In this contexts the parts of this collection “Science and new information” and “New horizons in cognition” should be approached. The articles under these headings range from reports of the concrete findings to philosophical and metho­ dological approaches to some issues. The former kind is re­ presented by the article “Man’s magnetic fields” by V. L. Vvedensky and V. I. Ozhogin who demonstrate how the measurements made by super-sensitive magnetic instru-

14

V. D. Pekelis

merits facilitate medical diagnostics and the studies of human brain. Yu. V. Gulyaev and E. E. Godik, the authors of the article “The physical fields of biological objects”, report a series of profound investigations accomplished by most advanced experimental methods, using precise physical instruments interfaced with computers. This work is being carried out now at the Institute of Radio Engineering and Electronics of the USSR Academy of Sciences. The complexity of topics discussed in this collection made it imperative to include philosophical-and-methodoligical articles. These are grouped under a general heading “Science and New Information”. The first article there, by A. M. Migdal, full member of the USSR Academy of Sciences, and Ye. V. Netesova, “Road to truth (on the scientific method of cognition)” discusses the differences between scientific and unscientific treatment of the phenomena in the world around us, between what is true and what is not in science. D. I. Dubrovsky (“Criteria of existence and conflicting situations in science”) analyzes from the materialistic point of view the criteria of existence and inexistence. In terms of knowledge and ignorance (we know that we know; we know that we do not know; we do not know that we do not know; which reminds of a biological catchword, from false knowledge to true ignorance) the author tries to analyze philosophically what is possible and what is not in science as far as informatics and cybernetics are concerned. B. V. Biryukov (“On new knowledge in biological studies”) defines scientific, pre-scientific, and unscientific sensations. The scientific process is analyzed by S. E. Shnol. His article “The dynamics of new truths in biological sciences” unravels the causes for delayed recognition of some disco­ veries. With all their variety the articles in this collection pursue

Note from the Compiler

15

the theme of the functioning of living matter and informa­ tion, or data processing and cybernetic aspects of biology. It is because many areas remain unexplored and problems unresolved that the truth and falsity in science and sensa­ tional findings had to be discussed. As biology employs data-processing methods on an ever increasing scale, new results are obtained and, which is equally important, well-known fact., are viewed from new standpoints. The ever-growing cooperation of cybernetics and biology facilitates the development of non-conventional ap­ proaches. Numerous new mathematical systems serve purely biological purposes. The application of informatics to living nature has yield­ ed certain negative results valuable in the sense that certain things were found impossible. Thus the functioning of the whole cannot be studied in a number of cases until “full” knowledge of its components is available. The super-com­ plex living systems incorporate two-way links which cannot be found unless the functioning of the components has been explored both “horizontally” and “vertically”. The data processing approach is very promising; thus far it has helped determine new aspects in the functioning of complex living systems. In the articles of this collection man is viewed as a com­ plex system. In studying man, every discipline should main­ tain its specific features and at the same time recognize the interaction of various fields of knowledge and the complex­ ities of comprehensive studies. The awareness of this aspect is obvious in all articles of the collection and helps the au­ thors either to answer questions or to pose new questions. Karl Marx foresaw comprehensive studies of man. He said: “Natural science will in time incorporate into itself the science of man, just as the science of man will incorpo­ rate into itself natural science: these will be one science”.

16

V. D. Pekelis

Indeed, now man is being studied by philosophers and socio­ logists, specialists in ethics and pedagogy, physiologists and medical scientists, and many other scientists. Quite naturally, all the aspects of these studies could not be treated in the framework of several Parts of this collec­ tion. The authors, however, tried to raise most interesting questions. In the context of the technological revolution and its consequences man becomes the strategic goal of cognition. This is what the article “On systemic, integral nature of man” is about. Its author V. G. Afanasyev, full member of the USSR Academy of Sciences, is one of the founders and a most active protagonist of systems research in the Soviet Union. He has made an important contribution to research on the philosophical dimension of the systemic character of nature and society. Man is said in his article to be not only the “centre”, or “focal point” of a social system but also a biological being which is a mobile self-controlled integral system, a concen­ tration not only of social relationships but also of the object­ ive world in the variety of its manifestations. In fact, man organically combines all laws of the universe, mechanical, physical, chemical, biological, and social, the latter being dominant and system-generating. Marx also said that man is a part of nature. But Marx and Engels provided a profound understanding of man’s dependence on nature and the social significance of that dependence. Marx wrote of human nature as a totality of his life powers. This essential formulation is in no way in conflict with another Marx’s formula, that man is a totality of all social relationships. A sound informatics-and-biological approach organically combines physico-chemical tools, a biological statement of the problem, and cybernetic ideas. In this way a new step

Note from the Compiler

17

is made towards understanding the phenomenon of life in the cognition of man and adds to the hierarchy of models which provide an insight into the super-complex systems of intelligent living beings. The article “Transition to constructing living systems” is worthy of special attention. Its author, A. A. Baev, full member of the USSR Academy of Sciences, is active in mo­ lecular biology and genetic engineering. He demonstrates that with the advent of genetic engineering the experiment­ al biology entered a new stage of development which can be regarded as very promising. The author comes to a con­ clusion that the concepts of information, coding, control, and feedback and of the entire control engineering methodo­ logy have enhanced the treatment of numerous conventional biological issues and, which seems especially significant, have been instrumental in restating some issues (such as decoding of the genetic information), and in resolving com­ plex problems in genetic engineering. Although its achievements are still modest, the promise of genetic engineering in the designing of living beings is such, notes Baev, that one has to restrain one’s imagina­ tion in forecasting the future successes. The Part “Brain and Intelligence” in this collection includ­ es a slightly abridged version of a widely-known article by P. K. Anokhin, full member of the USSR Academy of Sciences, a Lenin Prize laureate, “Natural intelligence ver­ sus artificial intelligence: the philosophical view” which still remains relevant. The author tries to show the limits of modeling the mind and the potential of developing artifi­ cial intelligence. Speaking of the cognition of the brain, its functional mechanisms, and molecular nature, Anokhin subtly leads the reader to a conclusion that the techniques of brain acti­ vities should be employed in designing artificial intelli2—0913

18

V. D. Pekelis

gence. The highlight of the article is in that the evolution of neurophysiological mechanisms suggests the systemic nature of their formation aimed at predictive reflection of the events in the environment. In characterizing the “func­ tional systems”, Anokhin indicates the difficulties inherent in a direct application of neurophysiological findings to control hardware. Reliability of the brain is the subject of A. B. Kogan’s article. An attractive feature of his article is the stringent boundary of his research. Another Part of the collection, “Novel Aspects”, reports interesting results in studies of biorhythms by V. N. Reushkin and of the biology of sleep by V. S. Rotenberg. In his article “Diurnal rhythms and adaptation”, Reushkin reports that studies of nearly-diurnal rhythms have experimentally shown that if an exogenic Signal which invokes an anxiety response (in H. Selye’s terminology) is repeated daily, then an expectation response is generated. This response, the author believes, improves the stability of the organism to a similar signal and improves its adaptability. Analysis of nearly-diurnal rhythms gives a clue to an individual’s health and makes it possible to predict illness. Sleep is known to be widely interpreted as a factor en­ hancing adaptation to the environment. A new explanation of this adaptive significance of sleep is offered in the article “Searching activity, sleep and stability of the organism” written by V. S. Rotenberg. His own research, V. V. Arshav­ sky’s experimentation, and analysis of the extensive litera­ ture on the subject have led Rotenberg to a conclusion that a kind of information-seeking activity takes place in one’s sleep. This activity offsets the biologically harmful ef­ fects of abandoning the search for necessary solutions in wakefulness. This mechanism may be regarded as a kind of feedback model which connects in one loop the psycholog-

Note from the Compiler

19

ical state of a personality, the physiological parameters, and the biochemical processes in the organism. The same Part of the collection includes an article by V. I. Klimova “On man’s third state”. The organism’s state when one is “neither healthy nor sick” is viewed in the light of data processing principles applied to medicine and bio­ logy. An increasing amount of attention is given to “the third state” because, unfortunately, too many people stay in it for too long. Several articles discuss psychological fields such as the psychology of cognition, subconsciousness and superconsciousness, and some problems of engineering psychology. Cybernetic techniques make it possible to study psycholo­ gical processes in conjunction with physical, biological, and social phenomena, to find some common features, and to de­ monstrate the specifics of psychology itself. Cybernetic tools have a major role to play in psychologic­ al research. New industrial processes and automatic systems make quite sophisticated requirements to man. Psychology has therefore to rely ever more heavily on mathematical modeling and computers. Although psychological processes cannot be downgraded to physical and physiological ones, this new line of research, especially in experimentation, can boast of significant breakthroughs. The cognition psychology has made important conclusions from the study of mentality and behavioural studies in gene­ ral psychology and kindred disciplines and then started theoretical and experimental analysis of mental processes. This new line of research deals with the construction of the perceived pattern, the forms and structure of human knowledge, and the relationship of automatic and conscious­ ly controlled cognitive processes. Numerous models of perception, attention, memory, and thinking processes are now available. On the other hand, certain difficulties arise 2*

20

V. D. Pekelis

from overestimation of the similarity between the functional structure of cognitive processes in man and the data proces­ sing structure in computers. In his article B. M. Velichkovsky analyzes the history, problems, and promise of the cognitive psychology; he believes that the approaches which neglect the complex, systemic qualities of the mind have an essentially limited application. The complexity of the sphere of the unconscious in human mind is the subject of the article “Subconsciousness and su­ perconsciousness” by P. V. Simonov, Corresponding Member of the USSR Academy of Sciences. The latest findings of basic research suggest that there are several levels of cons­ ciousness, in particular, of superconsciousness which follows its own rules; that man is not fully aware of the needs which dictate his actions; and that consciousness, subconsciousness, and superconsciousness interact in a certain way. The basic conclusion of the article is that a materialistic solution to the most urgent problems in science of man cannot be ob­ tained unless the most important functions of unconscious­ ness processes are recognized and these processes are classi­ fied into sub- afid superconscious ones which are essentially different. “The principle of active operator in engineering psycho­ logy” is described by B. F. Lomov, Corresponding Member of the USSR Academy of Sciences, who convincingly shows that the operator in a man-machine system cannot necessar­ ily be described as a mere communication channel. He illustrates his point with reference to a “pilot-aircraft” system which is very sensitive to the weak points in data exchange between its components. The final Part of the book is “The Organism and Age”. This topic has never been more relevant. The Earth popu­ lation is known to “age”. In 1950 a mere 200 million people,

Note from the Compiler

21

or 7.7 per cent of the world population, were older than 60. In 25 years there were 350 million of them, or 8.5 per cent. Every day 200,000 persons reach this age. The twentieth century is quite justly referred to as the age of increasingly long life. Demographers say that very soon the average life span will grow to 85 or even 90 years. The UN forecast predicts this too. Within the time span of 75 years, from 1950 to 2025, the number of people over 60 is expected to grow five-fold and of people over 80, seven-fold. Consequently, whereas in 1950 only one of every twelve people was over 60, in 2025 this will be true of one of every seven inhabitants of the planet. In the Soviet Union, where the average life span is very high, over 70 years, about three million people were recently reported to be older than 80, of which 300,000 older than 90, and over 20,000 older than 100. Whereas in 1941 there were only 200,000 old-age pensioners, in 1982 there were 35,000,000 of them. By 1990 the country’s population is expected to include nearly 50 million old and very old. The articles in this Part of the collection are, however, by no means concerned with the demographic aspects of this phenomenon. Rather, they discuss the physiological aspects, the working of the genetic program which the organism abides by. Is the self-regulating system of life faultless? What are the errors in life control function? And how does the organism counter these errors? What is very important, is it possible for man to know the genetic program and amend nature by skilful interference so as to make the old age active and healthy? The contributions to this Part of the collection describe a comprehensive approach to studies of the old age, the most interesting experimentation in the Gerontology Institute

22

V. D. Pekelis

of the USSR Academy of Sciences, and the nearly realistic “overhaul of man” in the future. The few articles on the cybernetics of life can by no means exhaust all the topics in this field. This collection will hopefully serve a more modest purpose of giving the reader a taste of advanced biocybernetic and physiological model­ ing, and of psychological studies. V. Pekelis

I. Biology and Information Biology Today Basic Tendencies in Physico-Chemical Biology YU. A. OVCHINNIKOV

It is not unusual to hear or read that the 21st century will be the Age of Biology. This promise is certainly debatable because it assigns secondary importance to truly momentous achievements in physics, mathematics, and chemistry, in engineering and in other fields of knowledge; nevertheless, it would be difficult to argue against the salient fact, namely, that the recent discoveries and accomplishments of biology are revolutionary in spirit and epochal in their scale and import. The scope of interests and problems in today’s biology is extremely wide: from elementary processes in the living cell to the development of the entire organism, to its in­ teraction with other organisms and the environment in the ecological system. Today’s biology is a rapidly evolving branch of science, rich in exciting problems and prospects, commanding an army of enthusiasts, and armed with the most advanced techniques and equipment. Biology holds key positions in solving the global problems mankind confronts, be it the battle against fatal disease, the food crisis, or the pollution of the environment. Biology is progressing very rapidly, but the rate at which one of its disciplines is moving ahead is incomparably high. This discipline, which matured in the 1950s, is the physico­ chemical biology.

24

Yu. A. Ovchinnikov

The “Visiting Card” of Physico-Chemical Biology The birth and development of this branch of the biological science inaugurates a new era in the investigation of living matter. This event, being one of the most remarkable in the history of natural sciences in this century, is in the spectacular penetration of the ideas and techniques of phys­ ics, chemistry, mathematics, cybernetics, and other such fields into biology. This swift breakthrough into the world of fascinating biological structures and giant molecules with their unique properties was possible due to the superior power of human mind and the tremendous potential of modern technical means. Physico-chemical biology is perhaps the first scientific discipline in which man recognized for the first time the unique dynamic architecture of the higher form of matter and was able to shed light on the extraordinary mechanisms that serve to ensure high efficiency, precise coupling, self­ regulation, and reliability of the living cell systems and of the entire organism. A new qualitatively different stage has begun in our materialistic interpretation of the living nature, even though nature only partly unveiled its mysteries. This is the stage of direct analysis of the most profound biolog­ ical processes. The era of physico-chemical biology is characterized by the development and rapid progress of a family of inter­ related scientific disciplines which draw upon the accumulat­ ed experience and employ the achievements of all modern branches of science. The writing on the “visiting card” of physico-chemical biology is the union of biochemistry and biophysics, molecu­ lar biology and biological chemistry; indeed, physico-che­ mical biology came to life as an interdisciplinary science at

Basic Tendencies in Physico-Chemical Biology

25

the hottest points of contact between these four disci­ plines. The rapid progress of the science studying the living matter was brought about by the joint effort of the sciences deal­ ing with the physico-chemical life-sustaining processes. The world of biological molecules, both small and gigantic, evolves as a self-consistent system with clear-cut distribu­ tion of roles played by individual subsystems and elements and with dynamic relations between information, physio­ logical, and functional processes. Today’s biology is already able to explain such immensely complicated life-sustaining phenomena as transfer of hereditary information, release and transformation of energy, transport of compounds and ions, propagation of nervous impulse, and many others. The explanations it supplies meet the niost stringent criteria imposed by physics and chemistry. Learning the Structure to Know the Function

One of the central problems of physico-chemical biology is to decipher the structures of biologically significant com­ pounds which participate in biochemical transformations within cells. Indeed, the capacity to carry out a specific biological function was encoded by evolution into the struc­ ture of biologically active substances. The way to compre­ hending the living matter is to find out its structure. This path is full of thorns, it demands time and devotion, so­ phisticated methods and equipment. Actually, all efforts and expenses are justified because the deciphering of a structure paves the way to understanding most complex phenomena and events, and the mechanisms that make them possible. Suffice it to recall the epochal significance

26

Yu. A. Ovchinnikov

for molecular biology of the deciphering of the double-helix structure of the deoxyribonucleic acid molecule (DNA), or of the determination of the amino acid sequence of insulin which was the starting point for the work on protein struc­ ture. Scientists are attracted to various levels of the structural organization of the living matter, but mostly to the structu­ re and spatial arrangement of biologically important mo­ lecules and the mechanisms of formation of molecular com­ plexes and ensembles. We obviously need to know how these structures change in time, i.e. to know the dynamic para­ meters and to have a precise kinematic description. Finally, it is necessary to discern the relationship between the struc­ ture and the biological function it represents. These prob­ lems cannot be solved unless the researchers mastered the whole spectrum of structural analysis techniques which of­ ten involve a cybernetics approach to problem description, complete automation of the experiment, and employment of a computer. Structural analysis invariably calls for months or years of intense labor; the romantic world of daring hy­ potheses becomes accessible only after the obtained concrete data have been decoded, and when imagination is fuelled and steered by the results supplied by physics and chemistry. In fact, the scope and sophistication of structural analysis characterize the maturity of physico-chemical biology, the basic truth, and the reliability of its concepts and conclu­ sions. By now the primary structure (the amino acid sequence) of hundreds of simple and complex proteins has been deter­ mined by the joint effort of scientists in a large number of countries. These very important biopolymers are responsible for all principal functions of organisms. Soviet scientists significantly contributed to “uncovering” protein structures. Notably, one of the first deciphered structures of ribonucleic

Basic Tendencies in Physico-Chemical Biology

27

acids (RNA) (that of valine) was determined in the USSR. The improvement of express analysis techniques for deci­ phering the nucleotide sequence of DNA, leading to a steady determination of DNA structures is another complex and difficult task. At present we know the sequences of a number of large DNA fragments (of several thousands of nucleotides each) and of several viral DNA molecules. For example, the Shemyakin Institute of Bioorganic Chemistry and the In­ stitute of Molecular Biology of the USSR Academy of Scienc­ es reported the nucleotide sequence of an important regula­ tory fragment of the DNA of one of bacteriofages. The length of this fragment was about 1300 pairs of nucleotides. The work on deciphering DNA structures unravelled a good deal of surprising results in the mechanisms of record­ ing and transmitting information in biological processes. For example, it was found that genes do not necessarily form a sequence: one fragment may belong to several overlapping genes. Furthermore, a sequence of nucleotides in DNA genes is not necessarily a continuous code for the sequence of amino acids in a protein, since parts of the sequence that codes for the protein can be separated and found in different parts of DNA. The flux of information on DNA structures remains vigor­ ous and keeps filling more and more pages in the Book of Biology. In this connection we should praise the substantial achieve­ ments in the study of the spatial structure of biopolymers and bioregulators. The structures of a large number of biologically important compounds have been deciphered by using X-ray structural analysis, nuclear magnetic resonan­ ce, UV and IR spectroscopy, and other modern high-pre­ cision techniques.

28

Yu. A. Ovchinnikov

Genetic Engineering Revisited The problems of genetics, which is defined as the science of inheritance and variability, are among the best investigat­ ed problems in physico-chemical biology. It is a well-known fact nowadays that the “genealogical tree” of each organism is written into a giant DNA molecule as a specific sequence of nucleotides. Relatively short messenger RNA molecules are synthesized on the DNA matrix; the process is catalyzed by enzymes. Special organelles of cells, called ribosomes, make use of these RNA sequences to assemble the appropria­ te proteins. Each property of a living system corresponds to a specific protein. Soviet scientists made a very significant contribution to the study of the mechanisms of storing and transmitting the inherited information. Thus they discovered important features in the structure of the genetic machinery of micro­ organisms and some higher organisms, analyzed in detail the mechanism of RNA synthesis, and discovered informosomes, i.e., the complexes of messenger RNA and proteins; now the structure and functions of ribosomes are being studied, and techniques for isolating individual genes v^jjrked out. The physico-chemical approach in genetics opens new vistas for medicine, agriculture, and other applied fields. The greatest asset of modern genetics is the creation of genetic engineering some ten years ago. Now ^scientists at­ tempt to restructure the genetic machinery in the desired way, so as to design new genetic systems. Is there a point in “playing” with this “biological erector set?” There certainly is, and an important one. The methods of genetic engineering make it possible to decipher and clearly demonstrate the arrangement and the mechanisms of functioning of the genetic apparatus of cells,

Basic Tendencies in Physico-Chemical Biology

29

and to specify the function of each gene. Of no less import­ ance is the feasibility of creating in the future “tailored” organisms with properties ordered by the scientist (e.g. useful microorganisms), eliminating the inherited “defects” of plants and animals, and assisting in the treatment of hereditary diseases of man. Gloomy predictions for genetic engineering were voiced in the West. This attitude stems from the threat of growing, unintentionally, agents dangerous for man, as a result of manipulations with genes of microorganisms. Indeed, the highest skill and profound understanding of a problem are not sufficient for experiments in molecular genetics; the experiment must be run under certain strictly controlled and monitored conditions. Stringent guidelines for the genetic engineering research were enacted in a num­ ber of countries, including the USSR. If these constraints are met, the safety of the personnel and the environment are guaranteed. The objections raised aim at social and ethical aspects rather than scientific ones. It would be hardly pos­ sible or useful to fence in the development of genetic engi­ neering; rather, it should be made to improve the life of mankind. “Nervous Impulse” and “Ion Channel” Physico-chemical approaches have recently proved to be fruitful in such fields as studies of the nervous system and higher nervous activity, and the analysis of immunity, i.e., in studying the processes in which informational aspects play an important role. In the USSR this field is studied within the framework of two all-Union programs, “Nervous Im­ pulse” and “Ion Channel”, which are coordinated by the Interdisciplinary Science Council on the Problems of Phys-

30

Yu. A. Ovchinnikov

ico-Chemical Biology and Biotechnology sponsored by the USSR State Committee on Science and Technology and the Presidium of the USSR Academy of Sciences. This research comprises the study of biological membranes. Membranes surround a living cell, its compartments and its organelles, and create there the conditions different from those in the ambient medium. Membranes are mostly built of lipids and proteins; among the functions important for the organism, biological membranes are responsible for the transport of nutrients and ions into cells and out of them. Membranes confront and identify foreign agents, viruses, and drugs, sense the signals coming from the environment, and serve to release and transform energy. They participate in the transmission of nervous impulses, in the formation of res­ ponses to hormones, in creating intercellular contacts, and in numerous other processes. The electrical and chemical membrane mechanisms res­ ponsible for the generation and propagation of nervous excitation have been unravelled, and specific regulators of nervous impulse transmission discovered. As a result, meth­ ods of treating various mental and nervous system disorders were suggested. Toxins secreted by snakes, scorpions, and some sea organisms, capable of very selective interaction with the most sensitive parts of nervous systems, seem to be very promising objects for such studies. Unique compounds, called neuropeptides, were recently discovered in the brain of animals and man. These substanc­ es can function as regulators of sleep and memory, can cause and releave the sensation of pain, fear, alarm, etc. Rela­ tively simple chemical compounds thus participate in very complicated manifestations of the higher nervous activity, supplementing the electrophysiological mechanisms of inhi­ bition, excitation, and recognition. Some of these compounds have been isolated in pure form, their structures were estab-

Basic Tendencies in Physico-Chemical Biology

31

lished, and they were even synthesized in laboratories. These “chemical” aspects of brain functioning, which made a good number of older concepts obsolete, deserve greater attention on the part of neurophysiology which is too often “anchored” to traditional interpretations. Biotechnology The recent decade witnessed a sort of “boom” caused by the advent of the modern biotechnology. A highly mobile, ef­ ficient, and compact branch of industry has grown on the latest achievements of biological sciences, using, above all, the methods of genetic and cell engineering. Biotechnology is a field intensively persued in the USSR; the basic economic guidelines for 1981-1990 specially under­ line the importance of biotechnological methods for national economy. * Let us consider several examples of the potentials of bio­ technology. First, biotechnology can produce industrially such unique bioregulators, previously unavailable, as insulin, interferon, growth hormone, etc., for medicine and agriculture. Severe forms of diabetes, which affect tens of millions of people on the globe, are treated with insulin of animal origin. Since the animal and human insulins have somewhat different structure, the patients often suffer from severe allergic reaction to the “foreign” substance. Attempts to synthesize human insulin had to be abandon­ ed for reasons of prohibitive cost. The solution to the prob­ lem was recently indicated by genetic engineering. The in­ sulin gene was isolated from a human cell. This gene was inserted into the DNA of conventional colon bacillus, E. coli, so that the fermentation tanks of biotechnological plants

32

Yu. A. Ovchinnikov

became sources of the unique human insulin. Preparations of this “microbial” insulin are now being studied by medi­ cal practitioners in a number of countries, including the USSR. Interferon is an even more spectacular example. Inter­ ferons are natural antiviral proteins which are produced by an organism in response to a viral infection. In pure form these proteins are practically inaccessible. The best source of human interferon is donors’ blood. Actually, patients suf­ fering from viral infections require so much interferon that this amount would be impossible to obtain even if all people of the Earth became donors. A different approach had to be found. As with insulin, scientists turned to the “DNA industry” and cell engineering. Interferon genes were isolated in laboratories of a number of countries, including the USSR, and pioneer experiments on “inserting” it into E. coli were carried out successfully. The stage of direct chemico-enzymatic synthesis of the gene of human interferon has been completed in this country, and first batches were manufactured on an industrial scale. Work on deciphering, isolation, and transplantation of nitrogen fixation genes is also very promising. Some micro­ organisms, such as nodule bacteria, are capable of digest­ ing atmospheric nitrogen when in symbiosis with certain plants (notably, leguminous plants). If it were possible to transplant genes of this type into the genetic apparatus of other microorganisms and cereal crops, the problem of nitrogeneous fertilizers would be largely alleviated, producing a virtual revolution in agricultural production. This line of research is pursued at a number of research centers, in­ cluding the USSR Academy center in Pushchino. Techniques available to modern science make it possible to cultivate in special nutritional media not only popula­ tions of microorganisms but also those of plant and animal

Genetics, Evolution, and Theoretical Biology

33

cells. A complete plant can be grown, and biomass compris­ ing all components of a mature plant organism obtained, from a single plant cell under appropriate conditions. Physico-chemical biology and biotechnology are two close­ ly interrelated areas of modern biology, being its growth points and its horizons. It would not be an exaggeration to say that today the state of the art in biology and biotechno­ logy determines, to a great extent, the scientific and tech­ nological potential of our country. Our achievements in this field are truly impressive.

Genetics, Evolution, and Theoretical Biology N. V. TIMOFEEV-RESOVSKY

All* entirely new approach came to replace in the 20th cen­ tury the former physical picture of the world, the picture which is in fact embodied in the familiar Laplacian determi­ nism philosophically “adapted” in the Auguste Comte’s po­ sitivism. The current outlook has not been “officially chris­ tened” yet, and we refer to it as the quantum-relativistic standpoint, since it rests on the modern quantum theory and theory of relativity. Imagine the absolute Laplace-Comte determinism: every tiniest motion is prescribed by some “world’s formula” that we are unable to use either owing to our ignorance or for a lack of data. Correspondingly, neither the freedom of con­ science nor-the freedom of opinion exist:*indeed, any possib­ le correct proposition is already contained in that formula. It would be quite silly to write a paper; suffice it to request 3-0913

34

N. V. Timofeev-Resovsky

a mathematician to derive from the general formula (and several thousands of auxiliary formulas which help to use the main one) the statements that the present article offers to the reader. Determinism of this sort essentially renders any practical activity meaningless. Indeed, the society need not formulate any objectives, since everything had been recognized and predetermined by the universal formula. Obviously, man has no place in such a world. The new physical picture of the world is radically dif­ ferent: it does not preclude the freedom to plan our indivi­ dual, collective, social, political and economic activities, and also the freedom of conscience. This picture is one of the main achievements of natural sciences in this century, even if it is not recognized as such by everybody. The main achievement of biology was the development of genetics. * * * Genetics was born in the 20th century as a belated though vital link in the mechanism of evolution that the genius of Charles Darwin discerned more than a hundred years ago. Darwin indeed saw the principle of Election in nature, and this succeeded in laying the foundations of evolution theory. Darwin clearly stated in the title of his most im­ portant treatise, “On the Origin of Species through Natural Selection”, that the way to construct the theory is to apply the principle of natural selection to some “individual varia­ bility”, i.e., to a nondirected statistical process involving equally well the most general and the most specific tiny details of living organisms. In fact, that was what Darwin did. Unfortunately, nothing was known in Darwin’s time about the elementary material for selection. There was practically

Genetics, Evolution, and Theoretical Biology

35

no cytology, chromosomes were unknown, and the principal paper of Mendel was published later than Darwin’s book. It looked as if Darwin’s titanic research had no foundations: the theory of evolution rested on an uncertain variability of quite unknown nature. Mendel’s genetic rules were rediscovered at the end of the 19th and the beginning of the 20th century, in five different countries and with 19 different objects. All speculations about the spurious origin of these observations were, there­ fore, silenced and Mendel’s mechanism of heredity emerged as a general law of nature. Time was right for starting the construction of evolution theory. In this connection we should recall the illustrious school of American cytologists created by E. B. Wilson (it was the best among the German, British, and American scientific schools) whose scientists completed the first step of studying the cytology of meiosis, that is, of the maturation of sex cells, and the cytology of fertilization. In 1902 Wilson pub­ lished in S c i e n c e a short note where he drew the attention of the scientific community to the observation of his col-. leagues W. S. Sutton and C. E. McClung. Namely, they found that meiosis and fertilization were nothing else but a cytological confirmation of Mendel’s brilliant hypothesis on hereditary factors and gametic purity. There is a similarity in Darwin’s and Mendel’s scientific fates. In contrast to frequently expressed incorrect opinion, Darwin was not the author of the concept of evolution, which had been expressed much earlier by Aristotle, Linnaeus, and many other predecessors. His genius was in that he was the first to discover the principle of natural selection, a natural mechanism for the evolution of living things with time. Likewise, Mendel’s genius was not in discovering the law of heredity, although this opinion is frequent and incorrect. Before Mendel’s work, these laws were known to some pract-

36

N. V. Timofeev-Resovsky

ising selectionists. His genius was in conducting, for the first time in experimental biology, a set of well-conceived experiments, exactly evaluating the results obtained, and formulating the gametic purity hypothesis. The result was a clear and unassailable interpretation of the data obtained in the experiments with pea plants. Evidently, Mendel’s work and especially Darwin’s ana­ lysis can serve as a foundation of the future edifice of the theoretical biology. * *

*

At present, we are not yet equipped with a theoretical biology as compared to theoretical physics. The discipline which is sometimes referred to as such has been known since the 19th century as general biology. Textbooks on general biology, which later became classic, were written at the onset of this century. Those books were “General Biology” by M. Hartmann and “General Zoology” by A. Kuhn in Germany, a number of monographs by J.B.S. Haldane and J. S. Huxley in Britain, and an excellent treatise “Biolog­ ical Foundations of Zoology” by Vladimir Shimkevich in this country. None of these books are obsolete today (wrong are those who think Darwin is out-of-dat#, actually, read­ ing “On the Origin of Species” today is of greater use to every biologist than a booklet about Darwin’s work even if it has been written only half a year ago). Later L. Ya. Blyakher published an extremely good textbook of general biology, and C. Villee’s “Biology”, translated into many languages, proved to be extremely successful in recent decades. Theoretical biology does not exist—or did not exist until very recently—because we failed to find (at least until very recently) the general natural biological principles that would be comparable with the principles reigning in physics ever

Genetics, Evolution, and Theoretical Biology

37

since the 18th century. Apparently, only two such general principles can be identified at present in biology. The first such principle (it has been known for more than a century) is undoubtedly that of natural selection. Time and again arguments flare up on whether this Darwinian principle is valid or needs replacement, but the doubts never stand up to serious scrutiny. “Normal” biologists do not fight about natural selection. Perhaps, only mathematicians not steeped in biology can accept, and try to prove, that evolving nature could do without natural selection. Biology has at its disposal another general natural prin­ ciple, even though it is less well known so far than the prin­ ciple of natural selection. It became clear at the end of 1920s and the beginning of 1930s that whenever living things reproduce, creating their likes, we invariably find the replication of molecules. This understanding was achieved by Max Delbriick and later by Paul Dirac (one of the members of the famous Copenhagen club of physicists and mathematicians clustered around Niels Bohr) on the basis of the physico-chemical model of chromosomes and genes developed by N. K. Koltsov. In contrast to crystal growth, which also involves the replica­ tion of molecules, the process unique to living matter is cal­ led r e d u p l i c a t i o n . One of the main manifestations of life is the growth in the number of elementary individuals, rather than in the mass of the living matter. In this process and elementary living being assembles its like and rejects it, thus launching into the world a new individual. Reproduc­ tion is not an adequate term for this process, reduplication being a much better one. After the advent of genetics in the 20th century it became clear that all organisms are subject to 3 spontaneous mu­ tational process, that mutations are inherited, and that re­ duplication passes them on to subsequent generations. In

38

N. V. Timofeev-Resovsky

our discussions with Delbriick and Dirac about the possible formulation of a general biological principle revealed in this process, we came upon a phrase which seems to be very convenient, namely, c o n v a r i a n t r e d u p l i c a t i o n , or reduplica­ tion of living entities which includes inherited variations. It became clear that convariant reduplication of discretely organized codes of genetic information is likely to represent the second general biological natural principle. So far this formulation is neither sufficiently rigorous no quite pe feet. Nevertheless, even at this stage it is evi­ dent that two general biological natural principles are estab­ lished. One is the natural selection, and the other can be christened, in a tentative fashion, the principle of convar­ iant reduplication of discrete codes of hereditary informa­ tion which is transferred from generation to generation. * * * In my opinion, it is justifiable to offer for discussion an additional biological phenomenon which is very promising from the standpoint of formulating a third biological prin­ ciple. This phenomenon concerns the problem of so-called p r o g r e s s iv e e v o l u t i o n . As a concept, progressive evolution still lacks not only a rigorous or exact, but even a minimally acceptable, rea­ sonable, logical definition. So far biologists did not deign to put in words what progressive evolution is. I think, the question is, whether the natural selection operating for a long time necessarily results in progressive evolution or not. Here, a full-size mathematical problem arises in biology. Until now biology gained very little from most of mathematical biology, or biological mathematics. In fact, skillful manipulation of mathematical formulas

Genetics, Evolution, and Theoretical Biology

39

fails to add to a profound understanding of the core of bio­ logical processes. I recall an interesting illustration. At the end of 1920s and the beginning of 1930s I took part in the development of the fundamentals of the modern physico-chemical form of the interpretation given to the principles of hit, target, and amplifier in radiobiology. A group of scientists at the German Institute of Metal Physics was interested at the time in applying mathematics to radiology. About 20 short papers were published, each containing about 20 formulas that were hardly comprehensible to biologists. I was partly responsible for attracting to this work first Max Delbriick, a student of M. Born and N. Bohr, who was originally a “pure” physicist and mathematician, and later Werner Hei­ senberg. After roughly a year of meetings of our colloquium, we achieved profound understanding of phenomena and of the description of processes, so that in subsequent publica­ tions the number of formulas dropped from 20-25 to 2-3. JThe famous French mathematician Henri Poincare once said—and later I heard Bohr say the same—that if a scien­ tist does not understand a problem, he writes numerous for­ mulas, but when he reaches understanding, at best two for­ mulas are retained. I believe that the problem of whether the natural selec­ tion, which has an eternity to do its job, necessarily re­ sults in progressive evolution (we prefer to think that it does) can be solved by an outstanding mathematician, or a group of outstanding mathematicians, who are at the same lime true philosophers. It seems that special mathematical methods have to be found, which would give a more or less definite answer to the question formulated above. The an­ swer may equip us with a third natural principle which could be employed for developing the theoretical biology. I had played an active role in formulating the second principle.

40

N. V. Timofeev-Resovsky

As for the third one, no one of the contemporary scientists seems capable of giving a serious answer to the question about inevitability of progressive evolution in the course of natural selection. * * * Before handing this problem to mathematicians, biolo­ gists first have to give a definition of progressive evolution, and second, to clarify whether different types of evolution are possible. Evolution on our planet followed different paths. Thus, it led to the mechanism of the higher nervous activity typical of man, but it also produced a fascinating organization of social insects. Life on the Earth could be very different if the evolutionary victors, and in a certain sense “kings of beasts”, were these insects. For example, the notions of morality and heroism would be meaningless: a bee that stings and thereby perishes is not a hero because it was meant to behave in this manner and equipped with special tools and weapons. Neither would there form those notions of ethics and those sublime categories which exist and will exist as long as the Earth is populated by people having a freedom to choose and a freedom to make decisions. We, biologists, stand in need of formulating (defining) a number of concepts that are involved in the formulation of the general natural principles which are required for deve­ loping theoretical biology. This done, we will enter the period of elaborating most diverse general schemes for con­ structing a theoretical biology which would be something above mere “general biology”. Future will show whether additional general biological principles (on top of the three named above) are necessary. Nevertheless, the evaluation of the theory of evolution promises to become the first task of theoretical biology. When this job is done, biologists will be able to single out

Transition to Constructing Living Systems

41

and understand what conditions and what additional factors channel and shape the progressive evolution which is the product of natural selection.

Transition to Constructing Living System s A. A. BAEV

The interpretation of traditional problems in biology became more enlightening when the concepts of information, encod­ ing, control, and feedback were used, and the philosophy of cybernetics as a whole was applied to biological systems. An even more significant factor was the formulation and solution of new problems, such as the deciphering of the genetic code. The ideas of control engineering constitute an important component of research programs and methods in modern biology. Besides, observation lost the status of the predomi­ nant channel of gaining biological knowledge. Experiment found its way into biology, although sometimes it creates only very approximate models of actual situations. Biology, this traditionally descriptive science, was transformed into an experimental science. One of the most stunning discoveries, which led to the advent of a discipline called g e n e t i c e n g i n e e r i n g , was the product of experiment. This new branch of molecular bio­ logy opened up totally unexpected vistas in studying here­ ditary effects, but at the same time it led to numerous de­ bates on whether the outcome of genetic-engineering re­ search will be a blessing or a bane.

42

A. A. Baev

Genetic engineering aims at assembling man-made genetic structures, and ultimately, at growing organisms endowed with novel hereditary properties. This research turned ex­ perimental biologists for the first time into designers of living systems, who control the genetic information accord­ ing to a predetermined program. The dream of Middle Ages was to synthesize a homunculus, or a tiny artificial human being. Alchemists relied on black magic. We live in a different age, and connect our hopes only with moderate potentials of experimental biology. Ac­ cordingly, we impose limits on our dreams. Nevertheless, genetics engineers will soon grow, beyond doubt, into the shoes of chemical synthesis specialists who left a long time ago a variety of compounds prepared by nature and created a huge kingdom of man-made organic compounds. The achievements of genetic engineering are still modest, especially in comparison with the outlined research pro­ grams, and yet they are very impressive. Indeed, they show that contemporary biology is no longer satisfied with inter­ preting, or reflecting, the surrounding world of living things and man as its component; biology turns into a practical tool for changing this world in order to better satisfy the needs of population. Lessons of Genetic Engineering Genetic engineering can be defined as a system of experi­ mental techniques which make it possible to assemble in vitro artificial genetic structure in the form of so-called re­ combinant (hybrid) molecules of deoxyribonucleic acid (DNA). A living cell is essentially a tiny chemical plant whose technological process is dictated by a hereditary program

Transition to Constructing Living Systems

43

written into one of its nucleic acids, namely, DNA. Physical­ ly and functionally, the program consists of blocks, or genes, each of which controls the synthesis of a specific product (typically, of a protein) and the execution of a specific function which is controlled by this product. The introduc­ tion of new genetic information into the cell by recombinant DNA molecules changes the “inherited features”, so that the experimenter obtains the organism adjusted to a specific task. It is an extremely complicated job to identify the required genes in DNA molecules (which are huge even in the simplest organisms) and then extract and assemble them into a func­ tioning structure. Tools suitable for this work are very subtle. These are enzymes designed by nature and contained in living cells. Some enzymes (called restriction endonucleases, or r e s tr ic ta s e s ) “cut” DNA molecules at very specific sites into longer or shorter segments, and enzymes (called l i g a se s) join them into a single chain. The synthesis of artifi­ cial genetic structures became a feasible task after such ferments were isolated from cells and purified. In outlining the chain of events leading to the rise of genetic engineering, it is necessary to emphasize that it did not spring Aphrodite-like from the foam and did not in­ troduce either a new approach to biological phenomena, or new cognitive ideas, or the need to throw out the traditional set of concepts (meaning current concepts, not those pre­ valent in 1940s). The understanding of inheritance mechan­ ics and the problems of this field remained the same but the possibility of penetrating deep into the phenomena was dras­ tically enhanced. It was as if the key to a locked door was found, and the research was free to advance. This does not mean, however, that new technology brought nothing new; on the contrary, completely unexpected discoveries were made at the very outset.

44

A. A. Baev

It was established that higher organisms, yeasts, and some bacteria have mosaic genes, i.e. genes which code for a specific protein but are interrupted by inserted segments (introns) which are not related to this protein. As a result of this structure of genes (quite typical for all higher orga­ nisms), the cell is the scene for the so-called processing which before this discovery was completely unknown. When the genetic information is realized, the gene is first copied (to­ gether with all inserted segments and informative sequences). This copy is called the precursors RNA (RNA stands for ribonucleic acid). Special enzymes then “cut out” all in­ serted segments and join the informative sequence into a “mature” messenger RNA. The corresponding protein is synthesized on this “edited” RNA molecule. As the next step, the nature of transposons of bacteria (mobile genetic structures) and of mobile elements of higher organisms was established. These are the two most impor­ tant results of recent years. The discoveries that the nearest future may bring can hardly be predicted. But we can be sure that discoveries are inevitable, and that they may force upon us the need to revise some firmly established dogmas. Can genetic engineering be used to create new organisms? This is a typical question. We have already answered it in the affirmative, but some qualifications are necessary. Cur­ rently experimenter manipulates with relatively small amount of information. Even if this amount were considerab­ ly increased, the information carried by artificially synthe­ sized structures could not be organized so as to produce a completely new organism. So far our picture of the structural foundations of genetic control is very incomplete. Only a limited number of genes can be introduced into a bacterium, whose genetic status is thereby changed to a rather limited extent. However,

Transition to Constructing Living Systems

45

the possibility of introducing into a cell an alien informa­ tion belonging to a different species, even a different type, changes the situation drastically. An E. coli bacillus carrying active human genes is definitely a new organism. A feature typical of genetic engineering is that the re­ production of some key genetic processes in the laboratory was realized at the molecular level. A function entrusted by nature to an organism as a whole has been turned in labo­ ratory conditions into an operation carried out at the level of a cell or a molecule. Experimenters treat genes without any mystic aura; for them a gene is a fragment of DNA either isolated from natural systems or synthesized in vitro. The recombination, or arrangement of genes into a new se­ quence, takes place in a glass tube, according to the choice and wishes of the experimenter. The role of the usually allpowerful random factors is then so constrained that becomes virtually negligible; the goal-directed activity of the scien­ tist, his professionalism, his art turn into major factors. This intrusion into a formerly forbidden field cannot help impressing the community greatly, all the more so because we witness the very first steps of genetic engineer­ ing and thus could not get accustomed to its fascinating promise.

Genetic Engineering and Technological Revolution The readers will remember that man employs biological pro­ cesses since time immemorial, for instance, for fermentation in producing bread, wine, beer, and other kinds of food. In fact, this was biotechnology mastered empirically a long time ago.

46

A. A. Baev

Science, above all microbiology, grew to form the founda­ tion of biotechnology in this century, or rather in its second half. Some achievements of biotechnology at this second stage of development are fairly well known. They include the production of fodder proteins from petroleum using yeasts, and the utilization of microorganisms for “extracting” enzymes, pharmaceutical products, vitamins, and so forth. It is less well known that plant cells serve to generate the active component of ginseng and some other physiologically active compounds. The rapid progress of genetic and cell engineering gave shape to what is now called “modern biotechnology”. This field is based on some fundamental results recently obtained mostly in the physico-chemical branch of biology. An exceptionally important role played by genetic engi­ neering in biotechnology follows from the unique method it offers for obtaining the required microorganisms and com­ pounds, namely, the colonization of the microbial cell by the appropriate genes. If the genetic structure introduced into the microorganism becomes stable, this method becomes efficient and cost-effective. Despite the similarity of the elementary biological processes in the original organism and in the organism modified by genetic engineering, the geneticengineering manipulations may result in final products that differ from the natural output compounds. Sometimes, na­ ture has no analogs of the designed compounds. The new biotechnology has another extremely important characteristic. Until very recently the isolation of natural mutants was the only way genetics could use to enhance the productivity of microorganisms. Among the set of mutants, only the strains with useful properties were selected. However, the natural frequency of mutation is extremely low, about one individual in ten million. Besides, the properties of mutants

Transition to Constructing Living Systems

47

are very seldom better than those of the predecessors. Ge­ netic engineering designs mutants according to a precon­ ceived plan and a chosen objective. Here lie both the speci­ ficity and the advantage. The possibility, at least in principle, of producing com­ pletely new types of plants is already a reality for genetic engineering. It is not yet clear, though, to what extent its methods will be effective for “designing” new breeds of farm animals. But even this prospect seems to be realistic, be­ cause part of the genes introduced into cells of animals were experimentally shown to be functionally active. The next generation of protective measures for agricultural plants is also an important subject for research. We mean the development of compounds which are found in nature and which control the behaviour of insects; such are the sexual attractants, or pheromones, which are analogous to the juvenile hormone. These substances can disorient pests, disturb the normal course of their development, and even completely eradicate them when combined with other com­ pounds. The advantages of the new plant protection methods are especially clear in comparison with fairly toxic and long-lived insecticides which have been employed in agri­ culture in the last decades. To date the most advanced branch of biotechnology is the microbiological industry. Its objective is well-defined: to achieve exclusively industrial production of fodders and of physiologically active substances, and in the future, of human food as well. Mankind will not terminate the tradi­ tional agricultural production but will supplement it with the microbiological one. This combination can supply high­ er-quality food than currently available, and at lower cost. Genetic engineering opens unlimited possibilities for the cooperation of biology and medicine. One of the uewer

48

A. A. Baev

problems is the treatment of hereditary metabolic disorders by transplanting to the patient his own cells carrying heal­ thy genes. Obviously, only monogenic metabolic deficien­ cies are meant. Of course, considerable experimental search is needed for solving such a complex problem. It is clear even now, that cell transplantation has a number of advan­ tages over the conventional transplantation. Namely, it is possible to obtain a viable cell from the patient, trans­ form it, and transplant into it any gene of the same spe­ cies. The new biotechnology is responsible for giving birth to an unusual branch of pharmaceutical industry: the “DNA industry”, which makes use of the processes based on genetic engineering technology. The method of recombinant DNA proved its unparalleled potential for the industrial produc­ tion of medically important substances, such as human in­ sulin, human growth hormone, vaccine against virus hepa­ titis, human interferon, and some others. The hope of producing diverse substances of considerable medical and commercial value by the above-described tech­ niques is becoming increasingly more realistic. Numerous obstacles undoubtedly lie on the way to converting the labo­ ratory procedures into technologically and economically attractive processes, but earlier experience points to the pos­ sibility of overcoming them. Such is the current stage in biotechnology which defi­ nitely manifests all signs of future growth. The next stage of development, even more ingenious and promising, is taking shape now. At this stage two mutually enriching lines of research are pursued: the study of the mechanisms of biolo­ gical reactions, on one hand, and modeling these reactions in order to synthesize simpler physico-chemical analogs, on the other.

Transition to Constructing Living Systems

49

Genetic Engineering and Social Atmosphere Situations which arise when new scientific ideas are inject­ ed into the social texture and into social practice may be very dramatic. A brief history of genetic engineering is quite instructive, even though it is not unique in this res­ pect. Once it was born, genetic engineering immediately attract­ ed the attention of scientists, press, and general public, but not by virtue of its scientific achievements (which at the beginning was very modest) but for reasons of quite differ­ ent nature. The issue was raised about the potential danger to mankind and environment, about the ethical and bio­ logical admissibility of a crude intrusion of man into na­ ture’s order. The opinion that genetic engineering is a threat to society is rather widespread. What are the signs of danger attribut­ ed to recombinant DNA? First, it is assumed that the im­ planted alien genetic information may transform innocuous microorganisms (such as intestinal bacilli, E. coli) into pathogenic organisms. Second, microorganisms containing recombinant DNA are assumed to be able to acquire unpre­ dictable ecological advantages and to shift the equilibrium of microbial populations in the environment. The first of these options seems hardly probable because even though pathogenic microorganisms are harmful to man, they are perfect creations in their own right, rather than a nature’s bungled job. Breakdown of ecological equilib­ rium appears more probable. Indeed, mankind proved very “proficient” in this respect, having disrupted, for example, the nitrogen balance and having introduced in the environ­ ment large amounts of pesticides which are required for the agricultural production but are fraught with harmful effects on the environment. Finally, it was conjectured that 4 —0913

50

A. A. Baev

microorganisms with implanted genes of some physiologic­ ally active substance, such as insulin, may colonize the gas­ trointestinal tract and result in treatment-resistant patholo­ gical states. In the long run, these apprehensions proved to be unfounded. The hypothesis of the threat of recombinant DNA was mostly generated in the American scientific community. These scientists published during the Gordon Conference in 1973 a sort of a manifesto which declared the potential danger of “hand-made” DNA and argued for a moratorium, i.e. temporary halt in all recombinant DNA research until the real issues become clear. Very soon, a campaign against the hypothetical menace of recombinant DNA was organized in the USA. Sensation-hungry press, radio, and TV immedi­ ately got into the act, so the stand taken by the scientists was given good publicity. That was the start. Let us emphasize that the anti-genetic-engineering cam­ paign was running strong only in the USA. In all other coun­ tries, including socialist countries, the reaction was more restrained and balanced. Later the American scientists broke into several groups. Some went on with their campaign as relentlessly as before. Others changed their attitude and engaged in concrete in­ vestigations aimed at assessing the danger posed by recombi­ nant DNA. The campaign has been terminated by now. Two factors contributed to this change: first, no experimental proof was obtained of accidentally arising danger of recombinant DNA, and second, genetic engineering proved its promise of the possibility of industrial application (outlined above). This argument was decisive for dropping the idea of rigid control in genetic engineering and ceasing to exaggerate the dangers. It is important, nevertheless, to evaluate the extent to

Transition to Constructing Living Systems

51

which science is prepared to face the dangers involved in recombinant DNA if it is proved to be harmful or grows to be dangerous in the future. The experience accumulated in working with pathogenic organisms for more than a century conclusively supports the argument that science is adequately equipped with knowledge and techniques for protecting the personnel, the population, and the environment. It must be recognized that the initially voiced apprehension was exaggerated. Obviously, this does not mean that the research and, to an even greater extent, industrial-scale ge­ netic-engineering technology can proceed without strict control and monitoring. A reasonable, sober assessment of the positive aspects and possible dangers of genetic engineer­ ing were predominant in the USSR from the outset. Concluding Remarks The advent of genetic engineering inaugurated a new phase in the evolution of experimental biology: its creative phase. Indeed, the biologist can now act as a creative personality, rather than a passive observer. As genetic engineering ela­ borates and refines its tools, the role it plays will undoubt­ edly increase; moreover, important unexpected break­ throughs in the understanding of the structure and function­ ing of the genetic system may occur in the nearest future. Both genetic engineering and the whole family of biolo­ gical disciplines usually joined under the title “physico­ chemical biology” are looking into the future with confi­ dence. The world of artificial genetic structures will gain the status of a legitimate child of science and technology, as the world of man-made compounds, synthesized by organ­ ic chemistry, did some time ago. We believe that the collect­ ive wisdom of mankind will prevent any antihuman uses of genetic engineering. 4*

52

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

In a wider context, the future of physico-chemical biology will be decided by man’s attitude to his environment. Should original nature be destroyed by man’s activities and the re­ sultant anthropogenic environment prove to be very differ­ ent from, sometimes, even opposing, the natural environ­ ment, then physico-chemical biology will turn into one of the main tools of reconstructing the surrounding world. If man follows the “habitual” path of preserving the natural habitat, physico-chemical biology will be successful in protecting the completeness and richness of the environment. Whichever of these two paths is taken by man in the nearest future, physico-chemical biology will serve him faithfully regardless of circumstances.

Autowaves: An Interdisciplinary Finding G. R. IVANITSKY, V. I. KRINSKY, and 0. A. MORNEV

Autowaves (“self-sustained waves”) is a generalizing con­ cept that was introduced into the field of waves and oscilla­ tions for putting in order the experimental data and theoret­ ical notions about the mechanisms of some important pro­ cesses observed in biology, chemistry, and physics. The simplest example of what nowadays is referred to as autowaves is the combustion wave. The advancing fire wall of a forest fire is familiar to mankind from time immemorial. However, it was found only recently that the propagation of autowaves governs such dissimilar processes as transmis­ sion of information in living organisms, contractions of the cardiac muscle, initial stages of morphogenesis in some pri-

Autowaves: An Interdisciplinary Finding

53

mitive organisms, processes of activation of catalysts em­ ployed in chemical industry, and many others. The interest in studying autowave propagation is stimulat­ ed by the following important fact: the breakdown of regu­ lar modes of propagation and the interaction of autowaves result in disorganization and chaos in systems controlled by such waves. Thus, disturbances of this type may lead to grave cardiac arrhythmias. The reason for separating autowaves into a special class of oscillatory processes is their clear-cut distinction from other types of waves known in science, e.g., electromagnetic and mechanical waves in liquids, gases, and solids. We know that wave motions in liquids are excited if certain energy is spent on the creation in the medium of the initial pertur­ bation which then propagates as a wave. In the final count, this wave propagates owing to large-scale mechanical mo­ tions, obeying of course, the law of energy conservation. Consequently, the wave initiated by an external perturba­ tion requires no additional energy for its propagation. This situation is naturally realized when we throw a pebble into the still waters of the lake: a part of the kinetic energy of the pebble is converted into the energy of the initial pertur­ bation during that short moment when the stone breaks the surface of the water. If the lake is shallow, the ripples still run out while the stone is already at rest on the bottom. The laws of propagation and interaction of wave pertur­ bations in such media (conservative media) are especially simple in the case of low-amplitude sine waves. Such waves go unobstructed through one another, their interaction being reduced to algebraic summation of oscillations at each point of the medium (the superposition principle). This behaviour explains, among other things, the formation of classical in­ terference patterns, i.e., a moire pattern composed of os­ cillating region (at the points where the amplitudes add up)

54

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

and quiet region (at the points where the amplitudes sub­ tract) of the medium (Fig. la). The same fundamental su­ perposition principle leads to the other two characteristic properties of waves: reflection from obstacles and boundaries, and diffraction (propagation around obstacles). The energy of the initial perturbation is indeed conserved in conservative media, but these media are not convenient for transmitting signals over large distances: in two- and three-dimensional media the energy density decreases as the distance to the source increases, and the shape of the signal is distorted by dispersion, i.e., the velocities of pro­ pagation are different for different spectral components of the signal. All of the above-listed properties are modified in an un­ expected manner as we go from waves in conservative media to autowaves. The table below shows that the only property common for the two types of waves is diffraction.

Fig. 1. Interaction between waves emitted by two sources (a) interference of waves on the surface of water; (b) autowaves in an active me­ dium do not interfere (colliding autowaves are seen not to propagate through each other &s in Fig. la, but to annihilate)

55

Autowaves: An Interdisciplinary Finding

Property

Conservation of energy Conservation of amplitude and waveform Reflection from obstacles Annihilation Interference Diffraction

Waves

Autowaves

+

— +

+

—

+ + +

+

Note. “Plus” sign indicates the presence of a property, while Uminus” indica­ tes its absence.

What then is the autowave? By definition, autowaves are waves propagating through active media, i.e., media with distributed energy resources. The simplest example of an active medium is the miner’s safety fuse. Here the energy (chemical energy) is stored in the powder core, and the autowave is the combustion front travelling along the fuse. As the wave propagates, the sub­ stance of the core transforms from the stable high-energy state (unburnt powder) to the low-energy state (ash and gases left behind the combustion front). Part of the energy released within the combustion region is dissipated, while the other part is consumed for priming the burning of the consecutive adjacent elements of the still intact segment of the fuse. The above example makes clear the following general de­ finition: autowaves are self-sustained signals which initiate the processes of local release of stored energy which is con­ sumed to initiate similar processes in adjacent regions. Autowave propagation resembles relay races: the signal is reproduced at each point of the medium and therefore travels through the medium without attenuation or distor-

56

G. R. Ivanitsky, V. I. Krinsky, 0. A. Momev

tion. The energy stored in the medium is not conserved in the process but is consumed for sustaining the signal; this explains the first two properties in the table. It is also clear why two colliding autowaves annihilate each other; indeed, the zone behind the wavefront of the travelling autowave in which the transition from high- to low-energy state takes place is the “burnt-out” zone (for the safety fuse, burnt-out in the literal sense of this word) where this transition has been completed. The oncoming autowave cannot penetrate into this zone, and thus two colliding waves annihilate each other (Fig. 16). The impossibility of either interference or reflection from boundaries and obstacles is explained by similar arguments. As for the wave deflection around ob­ stacles, i.e., the diffraction, autowaves are fully capable of it. This diffraction is explained here just as it is in optics, i.e., by Huygens’ principle. The Huygens’ principle for autowaves is formulated as follows: each point of the me­ dium, reached at a given moment by the wave front, becomes the source of elementary circular autowaves; with anni­ hilation taken into account, the envelope of elementary waves gives the position of the travelling wavefront at the next moment of time. This behaviour is illustrated in Fig. 2 which shows a standard construction according to the Huygens’ principle (Fig. 2 a ) and demonstrates how the autowave front follows a curved boundary of the medium (Fig. 26). So far we gave only one example of an autowave: flamefront propagating through a combustible medium and con­ verting this medium irreversibly into its “burnt-out” state. The so-called recuperating active media, in which slow pro­ cesses convert the medium from the low-energy state (after the passage of the wave) to the original state, have more in­ teresting properties. An important phenomenon is observed in such media: the formation of local self-sustained sources

57

Autowaves: An Interdisciplinary Finding

(generators) of waves of the revolving vortices type which completely change the regime in the active medium. The nature and spectacular properties of such vortices will be described a little later. Let us begin with three examples of recuperating active media: a burner with slow-supply wicks, a chemically active medium, and a cardiac muscle. It will

(b)

Fig. 2. Diffraction of autowaves (a) Huygens* construction; (b) autowave follows the boundary of the active medium (successive posi­ tions of the wave front are marked with numbers)

4 4 4 4 t

t

9

Z

be clear that the laws governing the propagation and inter­ action of autowaves in active media are independent of the specific physical realization. Imagine a burner designed as follows. Strips of asbestos are inserted into holes drilled close to one another in a me­ tal plate, the neighboring strips being in contact. The lower ends of asbestos strips are immersed into a batch of thick oil. Asbestos is nonflammable but serves as a wick when impregnated with oil. The rate of burning of the oil-impreg­ nated asbestos wick is higher than the fuel supply rate. The flame will therefore soon die out. Later, diffusion will re­ new the oil content in the wick, burning can be restarted, and the cycle will resume. The wick can thus be in one of three states: burning; pause (refractory period) during which oil saturates asbestos; and the quiescent state in which the wick is ready to burn. If we ignite one of the wicks of our

58

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

demonstration burner, it will ignite the neighboring wick. The first wick will soon burn out (oil will be used up) but the flame front will propagate further on. This is an elegant way of creating a recuperating active medium: in contrast to the safety fuse, each element (wick) can flame up not once but an indefinite number of times. Note that the flame can be initiated not only by an external source but also by the travelling flame of the autowave. This is achieved if the sequence of wicks is closed into a ring; the flame then goes in a circle. If wicks fill a two-dimensional plane, a revolving fire vortex (a spiral combustion wave) is formed. A chemical active medium was prepared by A. M. Zhabotinsky and A. N. Zaikin in 1970. It was a thin layer of a liquid in which the Belousov reaction occurs. In contrast to the majority of known oxidation reactions which pro­ ceed until one of the substrates (oxidizer or reducer) is used up, in this reaction a strong inhibitor is released which sup­ presses the reaction after only a small fraction of reagents is consumed. The inhibitor is then removed by slow proces­ ses, and the reaction can be restarted. This chemical medium is therefore active and recuperating, and the self-sustained oxidation wave can travel through it repeatedly, until sub­ strate resources (“fuel”) are used up. The Belousov reaction consumes about 1 per cent of the substrates per cycle, so that the oxidation wave can pass through the liquid about 100 times. In principle, the mechanism of oxidation waves is the same as that of combustion waves (burning is a parti­ cular case of oxidation): excited (“burning”) elements of the medium excite (“ignite”) neighboring elements. Of course, the most interesting among active media are those created by nature. The best known example is the nerve fiber. The impulse propagating along the fiber is actual­ ly an autowave, namely, an electrochemical wave of tran­ sition between two states: the quiescent state in which the

Autowaves: An Interdisciplinary Finding

59

potential difference across the membrane of the fiber is high (about —0.1 V) and the active state in which the po­ tential difference is low (about 0.02 V). When a nervous im­ pulse is sent, energy is successively released at each point of the membrane; its source is the stored energy of nonequilib­ rium concentrations of potassium and sodium ions on both sides of the membrane. Each nervous impulse has standard (for each cell) amplitude, length, and shape. In addition to one-dimensional active media (nerve fibers), there exist two- and three-dimensional media composed of excitable cells which function as nerve fibers do. Examples of such media are the brain and the spinal cord, nonstriated-muscle walls of intestines, womb and bladder, and also the cardiac muscle. Autowaves travelling through them have the same nature as waves in nerve fibers, differ­ ing only in impulse length and propagation velocity; how­ ever, they play very different roles in life-sustaining pro­ cesses. The impulse travelling along a nerve fiber transmits information while the excitation wave propagating, say, through the heart triggers a cascade of biochemical proces­ ses which initiate the contraction of the cardiac muscle; the regime of contraction immediately changes in response to any change in the propagation of autowaves. Populations of differentiating cells which exchange che­ mical signals, the retina of the eye, ecological systems, a number of electrochemical systems, and some others are also recuperating active media. A fact of decisive importance is that the principles go­ verning the functioning of all active media (physical, che­ mical, and biological alike) are found to be identical and can be described in terms of a language that avoids the spe­ cificities of a medium. An active medium is invariably a two-level system that can occupy one of two essentially different states: a high-

60

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

Fig. 3. Local sources of auto­ waves in chemically active me­ dia (a), (b) spiral waves revolving in a simply connected medium, (“rever­ berator”) (a), and around a non-excitable element of the medium (b); (c) source of concentric waves

energy and a low-energy states. When an autowave advances, the medium elements at the wavefront drop from the highenergy to the low-energy level. The energy released in this transition is consumed for triggering the same transition in the medium directly contiguous to the wavefront. The simplest active media (nonrecuperating media) remain at the low-energy level after the transition, so that no repeat­ ed propagation of the wave is possible (e.g., the miner’s fuse, or phase transition waves). In recuperating active media an autowave can propagate an indefinite number of times because slow processes of energy pumping return each ele­ ment of the medium to its high-energy level. In the case of the burner with low fuel-supply wicks the high energy level corresponds to wicks saturated with fuel,

Autowaves: An Interdisciplinary Finding

61

and the low energy level, to the wicks with fuel exhausted (but fuel is stored in the jar into which the ends of wicks are dipped). The high-energy state of excitable cells cor­ responds to a large difference between the potentials of the inner and outer sides of the membrane (—0.1 V), and the low-energy state, to a small potential difference (0.02 V). Elements of the medium are usually non-excitable until the process returning the system to the high-energy state is completed; the corresponding time interval is called the refractory time (this term came from the physiology of ex­ citable cells). Let us explain how revolving vortices, the so-called rever­ berators, arise in active media. We have already mentioned that reverberators are the most important wave sources: the introduction of reverberators into an active medium can entirely change the mode in which it functions. Figures 3, 4, and 5 give photographs of reverberators in active media of different types.

Fig. 4. Autowaves in atrium (a) electric excitation wave triggering the normal contractions of healthy heart; (b) rotation of the autowave during paroxysmal tachycardia (M. A. Allessie, K I. M. Bonke, F. J. G. Schopman). Heavy lines trace autowave fronts; numbers give time (in milliseconds) elapsed after the wave was started

62

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

Fig. 5. Spiral autowaves formed in the process of differentiation of slime mold cells (G. Gerisch)

Reverberators are naturally arising in inhomogeneous me­ dia in which wavefronts break (rupture) in the course of propagation. Because of the discontinuity (Fig. 6a), quies­ cent regions of the medium lie not only ahead of the wavefront but also on the side, in the vicinity of the point 0. As a result, at the next instant the autowave moves upward but also sideways, penetrating into the region to the right of 0; Huygens’ construction clearly shows how this happens (Fig. 6b). The envelope of elementary circular waves on the segment A B is clearly a straight line (the wave front has

Autowaves: An Interdiscfplinary Finding

63

been displaced, as predicted, upwards), while at the seg­ ment B C D it is an arc of a circle which terminates at the lateral boundary of the refractory zone at a point D . The incipient spiral is already apparent; finally it will coil into a reverberator. Note that as time goes, the circular segment B C D expands, making the point D slide along the lateral boundary of the refractory zone towards the upward-moving back side of the wavefront, until they meet at a point O' (Fig. 6c). The se­ cond stage in the evolution of the reverberator spiral be­ gins at this moment. The refractory “train” of the autowave to the left of 0' moves upward, while the wavefront immedi­ ately to the right of this point moves downward; hence, a discontinuity appears in the neighborhood of 0' right after the formation of the configuration shown in Fig. 6c. This discontinuity is immediately filled up by the wavefront revolving around O', by analogy to what we described above (see Fig. 6a, b). As a result, a complete turn of the spiral is formed after some time (Fig. 6d). As these processes are repeated, a reverberator with a greater number of wave turns is formed (see Fig. 3a or 9d). A reverberator is a fascinating wave source: it can sur­ vive in a “stand-by” medium having no elements producing

Fig. 6. Huygens1 construction at a discontinuity of the autowave propagating in an active medium (I. S. Balakhovsky) Arrows indicate the direction of wave propagation; a revolving spiral autowave, or reverberator, is finally produced

64

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

self-sustained oscillations. Furthermore, not being anchored to a geometrical feature of the medium (it is merely a revolv­ ing wave!), it can migrate in the medium. A spiral wave revolving around an obstacle (see Fig. 3b) has similar properties, but in contrast to a reverberator it is anchored to the inhomogeneity at which it was born, and thus cannot migrate. Another type of source of autowaves is a source emitting concentric waves (see Fig. 3 c ) ; such are self-sustained oscillators of the active medium, surround­ ed by nonoscillatory elements. Speaking of the properties of local sources of autowaves, we must not forget such a characteristic of a source as its topological charge. A source of circular waves (see Fig. 3c) is assigned zero topological charge. The topological charge of reverberators, however, equals the number of revolving spiral waves that form the vortex. In addition to vortices formed of a single helix (see Fig. 3a), “multibranch” reverbe­ rators are found to be possible; these vortices consist of several “branches” each of which is a helix revolving around the common center in the same direction. Photographs of ,such reverberators remind, to some extent, one of the pho­ tographs of galaxies. Compare a photograph of the M 101 galaxy from the Ursa Major constellation (Fig. 7) with multibranch reverberators in the Belousov-Zhabotinsky reaction (Fig. 8). By definition, the topological charge of a reverberator equals the number of its branches taken with a “plus” or “minus” sign, depending on whether the vortex revolves clockwise or counterclockwise. The topological charge of a multiple-branch spiral wave revolving around an obstacle is defined similarly. If a medium contains several local sourc­ es of autowaves of various types, it is justifiable to speak of the net topological charge equal to the algebraic sum of topological charges of all sources. The following conservation

Autowaves: An Interdisciplinary Finding

65

law always holds in experiments: interactions between local sources of autowaves do not change the net topological char­ ge of the system (Fig. 9). The mode of oscillation in an active medium containing several sources of autowaves is dictated by interaction pro­ cesses. The source with the highest frequency suppresses all other sources, owing to mutual annihilation of waves. Among all local sources of autowaves, the highest frequency is found to be that of the reverberator. Consequently, it

Fig. 7. Photograph of spiral galaxy M101 in the Ursa Major con­ stellation 5-0913

66

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

Fig. 8. Reverberators with various values of topological charge

forces its rhythm on the whole medium, suppressing, among other sources, all concentric-wave sources (Fig. 10). Another important property of reverberators is that they can proliferate. We have already mentioned that a reverbera­ tor “coils around” a discontinuity “inflicted” by an inhomo­ geneity of the medium. If at least one reverberator is creat­ ed in an inhomogeneous medium, the autowaves it sends out are broken on inhomogeneities, thus generating new vortices, etc. As a result of this cascade process the whole medium finally gets filled up with segments of revolving

Autowaves: An Interdisciplinary Finding

67

spiral waves; the resulting chaotic pattern resembles a welldeveloped small-scale turbulence. These two properties of reverberators produce disorgani­ zation of the normal functioning of biological systems. Is

Fig. 9. Conservation of topological charge in vortex decay in the Belousov-Zhabotinsky reaction (a) two vortices with opposite topological charges (the net topological charge N = 0); (b) a droplet of reaction inhibitor was added at the vortex center (seen as a dark spot); (c) vortices are destroyed, leaving behind a circular wave with the same topological charge (N = 0); (d) through (/) the same procedure cannot destroy a single vortex (the topological charge iV = 1 is conserved) 5*

68

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

Fig. 10. Annihilation of a source of concentric waves by reverberators As a result of annihilation by oncoming reverberator waves, none of the con­ centric waves is closed (reverberators have higher frequency); note that the next reverberator wave front will reach the source of concentric waves

it possible to prevent such failures and thus control rever­ berators’ functioning? It is found that a reverberator can al­ ways be “muscled out” of the active medium by sending a high-frequency sequence of autowaves generated by an ex­ ternal source. Such waves make the reverberator drift in the direction of wave propagation (Fig. 11). This phenomenon can be utilized for “blowing” the reverberator toward the medium boundary: the vortex decays because autowaves are not reflected at the boundaries of the medium. Let us consider examples of how different modes of auto-

Autowaves: An Interdisciplinary Finding

69

Fig. 11. Control of the location of a reverberator in an active medium (a) a high-frequency train of waves is sent onto the reverberator (from below); (5) the reverberator is transformed into two closely spaced discontinuities drifting upwards and to the right (c); (d) after the source of high-frequency waves is swit­ ched off, the reverberator is restored at a new location

wave propagation and the interaction between wave sources affect the functioning of some biological systems. We begin with the work of the cardiac muscle. Under nor­ mal conditions the contractions of this muscle are controlled by a special source of excitation waves, the so-called sino­ atrial node. This node is a group of excitable cells located in the right auricle and functioning in the self-sustained mode. Approximately once a second the sinoatrial node emits a circular excitation autowave (see Fig. 4a). The wave propa­ gates through the auricles to the ventricles, causing synchron­ ized contractions of these chambers. The regularity of heart’s contractions is disrupted, if autonomous vortex sources of autowaves (reverberators) are produced in it for one reason or other. Reverberators suppress the normal activity of the sinoatrial node and disrupt the regular rhythm of heart’s contractions. It has been shown recently that this phenomenon is indeed observed in paroxysmal tachycardia, which is a severe cardiac arrhyth­ mia caused by the circulation of a spiral excitation wave (see Fig. 46). Avalanch-like proliferation of reverberators in nonhomogeneous regions of cardiac tissue results in another severe form of pathology, namely, the ventricular fibrillation,

70

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

in which the synchronism of contractions of individual myo­ cardial cells is lost, and the heart turns from a living pump into a chaotically jerking muscle bag unable to pump blood; this effect is a cardiac analog of the turbulent autowave mode in the Belousov-Zhabotinsky reaction. Reverberators arise not only in the cardiac muscle. As early as in 1944 a study of epilepsy revealed the so-called spreading cortical depression (SD), or the wave of slow shift of the intercellular electric potential of neurons in the cere­ bral cortex. In the norm all cortical neurons exchange nervous impulses: the mind functions through the interaction be­ tween neurons. As a SD wave travels through the nervous tissue, the neurons first undergo an intensive discharge and then cease their activity, so that the functioning of the cor­ tex is suppressed. A train of SD waves is a typical response to a powerful salvo of neuron discharges* which may be caused either by external causes (electric or chemical dis­ turbance) or by purely internal ones (as may happen in an epileptic fit). The series may sometimes last for several hours. One of the mechanisms responsible for such a long train of SD waves is their reverberation. This mechanism was recently confirmed in elegant experi­ ments on chicken’s retina. Retina is in fact a piece of the brain located at the periphery; it is capable of sustaining SD waves as the neuron tissue of the cortex is. These waves modify the properties of the retina so that reverberators (when they arise) can be photographed (Fig. 12). * It is assumed that the SD effect is attributable to the accumula­ tion of an excess of potassium ions in intercellular spaces; neurons are supposed to eject the ions in the course of intensive generation of im­ pulses. The normal level of ion concentration is restored by energyconsumed ion pumps of the living cells which pump potassium back into the neurons.

Autowaves: An Interdisciplinary Finding

71

Fig. 12. Spiral autowaves in the chicken retina

It is noteworthy that in some cases reverberators assist in the creation of order in the initially disordered medium instead of generating “chaos” in it. This effect is observed in populations of slime mold (D i c t y o s t e l i u m d is c o id e u m ) so­ cial amebas, presenting an instructive biological illustra­ tion of the self-organizing system. The slime mold is a system which can exist, depending o n

72

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

the external conditions, either as independently surviving single-cell amebas or as a single multicellular organism. If the medium is sufficiently rich in food, amebas live and feed independently without forming a coherent common organism. A sharp metamorphosis occurs when the food store is depleted. Then some amebas which have crossed the “hunger threshold” begin sending control signals into the medium, by periodically ejecting portions of cyclic adenos­ ine monophosphate (cAMP) which acts as a chemical signal substance. When the cAMP “danger signal” diffuses through the medium and is detected by the receptors of other amebas, the latter turn round and start moving along the cAMP gradient, secreting it as they move. At the final stage of this process the whole population crawls together, individual amebas join up and form a single multicellular organism called plasmodium. This organism has rudimentary loco­ motion organs and starts moving in search of food (perform­ ing this faster than an individual ameba would). Having located a food store, the plasmodium “breaks down” into free amebas which resume their individual existence. If the plasmodium fails to find food, the reproduction pro­ gram takes over, ensuring the survival of the species: the plasmodium differentiates and forms a fruit body, that is, a stem at the end of which spores grow in a special bag. After maturation, the fruit body bursts and the spores scat­ ter around in order to live through unfavorable conditions and produce a new population of amebas. It is readily noticed that in the course of cAMP genera­ tion the amebic population as a whole behaves as an active medium in which each ameba, detecting cAMP molecules and secreting in response new portions of cAMP into the surrounding medium, acts as an active element reproducing the “danger signal”. In this context the cAMP concentration waves are nothing less than autowaves for which (as for any

Autowaves: An Interdisciplinary Finding

73

autowaves of other types) all the above-described modes, including vortex modes, are possible. Vortices are observed in the structure of motion of migrating amebas (see Fig. 5). Here again revolving reverberator vortices are the fast­ est of all local sources of autowaves, because autowave sour­ ces have identical properties in all active media, and all other sources are, therefore, suppressed. When a reverbera­ tor is formed, the amebas crawl precisely toward its center, and there the fruit body develops! This is an example how nature uses reverberators for building up a structure in ex­ treme conditions. The fact that autowaves propagating through various active media have common features and identical character­ istics of local wave sources (with the mechanisms generat­ ing these sources, the interaction between them, and their proliferation being identical and independent of the speci­ ficities of the medium) offers a unique possibility to extend the laws established for autowave behaviour in an active medium to a broad class of media of a different physical nature. This was indeed the case when revolving vortices (reverberators) were discovered: they were predicted by theoreticians who analyzed mathematical models of pro­ pagation of excitation waves in the cardiac muscle, and were later experimentally produced in a chemical active medium. Still later reverberators were found in independent expe­ riments with heart, with populations of slime mold amebas, and with chicken’s retina. Nowadays a transfer of ideas and results from one scien­ tific discipline to another has become an effective working tool for research into autowaves. There is much to be gained by a consistent application of this principle. The results ob­ tained in the studies of autowaves on the cardiac muscle, in the work on arrhythmia mechanisms and methods of controlling them are extremely important because they

74

G. R. Ivanitsky, V. I. Krinsky, 0. A. Mornev

directly address the life-and-death aspects of human exist­ ence. The organizing role of autowaves in morpho- and embryogenesis processes is not less interesting, as dealing with two central problems of biology; indeed, it is well known now that the propagating and “frozen” autowaves (the latter are also called dissipative structures) play a de­ cisive role in shaping processes not only for D i c t y o s t e l i u m d is c o id e u m amebas but also for more complex organisms, even vertebrates. Another pressing need arising in biology is the study of autowaves governing the propagation of epidemics, ecolog­ ical invasions, and destabilization of biogeocenoses. The practical importance of solving the problems they entail is self-evident.

Cybernetics’ Standpoint

Cybernetics Approach to Theoretical Biology A. A. LYAPUNOV

Biology, as we know it, has accumulated vast amounts of empirical data concerning the description of structure of living organisms, their ensembles, and life-sustaining pro­ cesses. In each biological discipline, the reigning stand­ points and objectives dictate the way to systematize the relevant information. At the same time, the effort aimed at systematizing the biological data as a whole from a uni­ fied theoretical point of view is clearly inadequate. This si­ tuation is probably caused by the copiousness of the data, on one hand, and by insufficient theoretical understanding, on the other. Nevertheless, some attempts are worth making at the moment. The fact is that cybernetics opens up new theoret­ ical possibilities and contains a promise that fresh unifying concepts will arise in biology. The aim of the physico-chemical approach to biology is to reveal the elementary life processes and to study them within the framework of the physico-chemical standpoint. The goal of the cybernetics approach to biology is to form a holistic understanding of life processes using the knowledge of the structure of organisms and the elementary life proces­ ses. A synthesis of these two approaches may, hopefully, lead to the birth of unified theoretical biology. As a first step, we need to define the domain to be con­ sidered and to formulate clearly the problem and the basic concepts.

76

A. A. Lyapunov

In the present paper we intend to outline a cybernetics approach to analyzing life phenomena. 1. Empirical natural sciences accumulate factual data and offer its primary systematization. The theoretical branch rearranges the data into a coherent whole and looks for general laws reigning in nature and revealed in the data. The mathematical branch constructs model objects obeying similar laws and studies their behaviour. The completeness with which the basic laws of a field of knowledge have been revealed is found by comparing the functioning of these models with real systems. This is how the experimental, theoretical, and mathemat­ ical branches of natural sciences are interrelated. In the case of biology, the empirical branch is very well developed, the theoretical branch is much less so, while the mathematical branch appears to date as a collection of loose­ ly related particular theories. 2. For the systematization of the empirical biological data in biology, it is necessary to work out a unified stand­ point, equally essential for all biological disciplines. The biochemical, or bioenergetic, concept, based on treating the physico-chemical processes which make up the founda­ tion of life, can serve as this unifying standpoint. Another possible approach is offered by cybernetics. This approach requires the study of control systems in living beings and of the control processes necessary to sustain life. In the future, these approaches will permit the construction of the mathematical models of life processes; a synthesis of the two will prove to be most fruitful achievement. 3. Specialized control processes constitute an essential feature of life-sustaining processes. The main characteristic of the former is that the transfer of small masses or small portions of energy results in processes which transfer or convert much greater amounts of energy or mass.

Cybernetics Approach to Theoretical Biology

77

The purpose of cybernetics (or rather, control engineering) is to analyze control processes and the structure of control systems by mathematical tools, so that it is quite natural to make use of this science for studying the control of lifesustaining processes. In what follows we shall operate with a system of exact concepts introduced by control engineering, namely, i n f o r ­ m a t i o n , c o n tr o l s y s t e m , e l e m e n t a r y a c t , s i g n a l , and c o m m u n i ­ c a tio n c h a n n e l. 4. Control via information transfer constitutes a compo­ nent of any life-sustaining activity; in fact, control can be said to constitute the characteristic attribute of life in the broad sense of the word. The possible counterargument that control is widely used in industry is groundless because machines are designed by people, i.e., by living beings en­ dowed with conscience. 5. Let us attempt to give a definition of life proper. Note that so far biology was unable to define the terms “alive”, “life”, and “life-sustaining process”. This situation does not involve any difficulties for the descriptive biology, but creates extreme complications for theoretical biology, and even more so, for mathematical biology. 6. We propose to choose a set of phenomena which is broader than that of phenomena of life, but one which is well defined, and also to take the set of the branches of knowledge which study this chosen set of phenomena. Mak­ ing use of some accurately described attributes, we will try to single out from this set those which are identifiable as manifestations of life. At the same time, we shall define the set of biological disciplines. 7. We begin with considering distinct states of matter and those fields of natural science which study these states. A state of matter can be described if we choose spatial and temporal scales and a set of physico-chemical characters -

78

A. A. Lyapunov

tics. This set of characteristics must satisfy the condition of sufficient completeness. Unfortunately, it is hardly possib­ le now to list the necessary characteristics. This list must include mass, total energy, free energy, chemical composi­ tion (in terms of elements, especially in terms of stable groups of atoms, or individual chemical compounds), and possibly the magnetic and electrical characteristics of bodies. Dif­ ferent combinations of the said characteristics may be necessary for particular problems. All these characteristics must be defined for a part of the analyzed substance within a randomly positioned sphere which is entirely buried in the substance to be analyzed. The law describing the distribu­ tion of positions of this sphere must be prescribed in advan­ ce. Very often the center of the sphere is supposed to have constant distribution density everywhere in the admissible domain. 8. We are mostly interested in the mean values and var­ iances of the chosen characteristics. Let us identify the substances which are characterized by relatively low varianc­ es at a given mean value of the characteristic. We refer to these substances as homogeneous. Note that the homogeneity of a substance essentially depends on the diameter of the chosen spheres. The variance of characteristics increases with decreasing diameter in any substance. 9. Now we consider the variation of the characteristics of the material studied in time. We are interested in mate­ rials which differ from other materials by a higher stability of their characteristics as functions of time. Consequently, it is necessary to choose a unit time, and to study how the chosen characteristics vary within one unit of time. Materials whose average characteristics remain almost constant in time as compared with those of other materials (having close values of the same characteristics) will be referred to as relatively stable.

Cybernetics Approach to Theoretical Biology

79

10. In general, the stability increases with increasing diameter of the spheres used to determine the characteristics. If the fates of different parts of a material are random and mutually independent, it is possible, under some very ge­ neral assumptions, to find the relation between geometric size and stability. Of interest are the cases in which the stability of large aggregations proves to be higher than pre­ dicted by the theory. The material is then said to have enhanc­ ed stability. 11. Two sorts of factors influence the stability of a sub­ stance. Stability may result from unusually favourable ex­ ternal conditions, such as preservation or thermostating. Such situations are of no interest here. Another type of sus­ taining the stability is traced back to the internal response of the substance to external factors, which tends to maintain the equilibrium. Response of this type is said to be preserv­ ing. Materials of interest for us are those which have preserv­ ing responses. 12. Preserving responses arise when the substance re­ ceives the information on the external factor, processes this information, and generates new information, namely, a physical system of signals which trigger an internal restruc­ turing of this substance, such that the main characteristics of the substance are preserved. 13. Both the input and the output information are en­ coded in a finite number of discrete signals which are al­ lowed to assume a finite number of distinct values. Each signal is realized either by a specific physical process or by a specific state of a material object. Carriers of signals change in the course of data processing. Systems of signals of one type are thereby transformed from one code into another. 14. The data is processed by a device of discrete nature, called c o n t r o l s y s t e m . A control system is composed of indi-

80

A. A. Lyapunov

vidual elements connected by communication channels. Some of the elements act as inputs and some as outputs. The function of communication channels is to transmit signals. The elements receive, process, store, and yield sig­ nals. The device used to store the data is called the memory. In most cases data is stored either in cyclic combinations (loops) of elements and communication channels in which signals circulate or in elements which are capable of occupy­ ing several stable states and of going from one another in response to the input signals. In general, the way in which an element processes the signal is dictated by the type of this element of control system and by the state which it occupies. 15. Information stored in the control system is encoded in the discrete states of a finite number of discrete compo­ nents of this control system, so that each of these compo­ nents occupies one of a finite number of allowed states; in other words, the information is stored as a text of finite length, written in an alphabet consisting of a finite number of symbols. The way in which the incoming data is processed is essen­ tially a function of the information stored in the memory of the control system. 16. The control system, whose function is to generate preserving responses to various external stimuli, receives information on these stimuli, splits it into components, and compares them with the information stored in the system. The output information is composed in correspondence with the results of this comparison. The flexibility of the con­ trol system depends, therefore, on the amount of information stored in the memory. 17. Hereafter we will be interested in a substance which is homogeneous only on a sufficiently large scale, is rela-

Cybernetics Approach to Theoretical Biology

81

lively stable, has enhanced stability, and possesses a control system which generates preserving responses. Now we need to analyze how information can be stored in such control systems. 18. External factors are classified by characteristic time. For instance, there are external factors due to the motion of surrounding bodies with a velocity of the order of several meters per second, external factors due to weather fluctua­ tions, diurnal and seasonal external factors, and finally, pe­ rennial factors. The response time of preserving response must be match­ ed with the characteristic time of external factors. This con­ straint imposes certain conditions on the response time of control systems, for example, on the time of information retrieval from memory. 19. In general, the diversity of external factors is con­ siderable. Responses are less diverse, because the same response is often a preserving one with respect to more than one type of stimulus; nevertheless, in some cases this diver­ sity is also substantial. One immediate consequence is the large amount of stored information. Hence, the control system must have large data storage. We conclude that the control system is capable of responding to the multitude of external factors if its memory operates with a large number of material carriers of information symbols. 20. The environment thus imposes two types of require­ ments on the control system and its memory, namely, sufficient­ ly short response time and storage of a large amount of data. Obviously, it is not easy to meet these contradictory re­ quirements simultaneously. The difficulties are partly alleviated by the fact that these requirements are to some extent anticorrelated. The diversity of the fastest external factors (among those which cause preserving responses) is relatively poor. The 6-0913

82

A. A. Lyapunov

longer the characteristic time, the richer the spectrum of external factors. We conclude that the requirements formulated above can be met by a set of control systems some of which are slower but have larger memory, while others are faster but have smaller memory. In reality this set of control systems will have dissimilar physical mechanisms of storing and trans­ mitting the data, different anticorrelated ratios of computa­ tional speed and memory size, and different physical prin­ ciples of operation. 21. Note that each response involves an actuator, or actuators, having certain power and mass. On the whole the efficiency of functioning is the higher, the greater the masses, power, and energy of the actuators. The concentration of energy stored in actuators is necessarily limited. Hence, to improve the efficiency of the whole, it is important that the control system has a relatively small volume. The information carriers must, therefore, be very small. 22. Let us discuss in more detail the control systems which generate preserving responses and at the same time meet the severest constraints on the geometric size of the carriers of information symbols. The storage of data in the control system’s memory must be extremely reliable, otherwise the information cannot maintain the stability of the whole. At the same time, structures formed by a small number of non­ interacting molecules can never be stable, in view of, for example, thermal motions. Stable materials serving as in­ formation carriers can, therefore, be either macroscopic (in this case their stability is caused by statistic factors) or monomolecular, or crystalline, but in the last case the in­ formation carrier is the type of crystal lattice composed of identical periodically arranged unit cells. This mode of information storage entails very high redundancy, and thus cannot be economical in volume utilization.

Information Theory and Evolution

83

23. The substance we single out is, therefore, bounded and homogeneous, relatively stable, has enhanced stability, produces preserving responses, and incorporates a control system which generates these responses and uses information encoded and stored in monomolecular carriers. We refer to this substance as living matter. In short, life can be defined as the highly stable state of a substance which generates preserving responses dictated by information encoded in the states of individual molecules.

Information Theory and Evolution M. V. VOLKENSHTEIN

The concept of information appeared in physics in the con­ text of developing the foundations of statistical mechanics, although the term itself was not yet used. The relation be­ tween entropy and the probability for a system to occupy a state, established by Boltzmann, implies the relation be­ tween entropy and the amount of information. Entropy is a quantitative measure of the lack of information about the system. The basic propositions of statistical mechanics are derivable from the canonical information theory developed by Shannon and some other scientists in the context of the problems of communication theory. The equivalence of entropy and the amount of information was first pointed out by Szilard; in fact, this equivalence indicates a simple con­ servation law: for a given probability distribution of state occupancy, the sum of microscopic information and entropy is constant and equal to the maximal obtainable informa­ tion or to the maximal entropy of the system. Obviously, both entropy and information must be expres­ sed in the same units, such as bits or units of energy divided by temperature. Increased information entails decreased 6

*

84

M. V. Volkenshtein

entropy, and vice versa. Here both quantitites are treated microscopically. The equivalence of information and entropy is no more surprising than the equivalence of mass and energy implied by Einstein formula m = c ~ 2E , where c = 3 x 1010 cm/s is the velocity of light; 1 erg is equivalent to a mass of 10”21 g. Likewise, 1 bit of informa­ tion is equivalent to k In 2 = 10 ~16 erg/K of entropy, which is a very small amount of entropy (A; = 1.38 x 10“18 erg/K is Boltzmann’s constant). The message of this equivalence is that new information is obtained “at a price” of increased entropy (in a different part of the system). No information can be obtained about the state of an adiabatically isolated system. In other words, some energy must be dissipated. The minimum energy con­ sumption per one bit of information obtained is k T In 2, where T is the absolute temperature. We speak of obtaining information without seeking for the profound implications of this process. The capabilities of information receptors are very limited in the standard canonical information theory used for developing the foun­ dations of statistical mechanics or for solving problems in communication systems. A receptor can only distinguish be­ tween distinct states and between letters in a message. This is an obvious advantage of the canonical theory. When solving a problem dealing with the number of telegrams that can be transmitted through a communication channel, the content of the messages is ignored. Physics is based on receiving information, i.e. on meas­ urements. Only quantities that are measurable in principle carry a physical meaning and contain information. We are justified in using the canonical theory when analyzing

Information Theory and Evolution

85

measurements only as long as we disregard their conse­ quences. These aspects were thoroughly analyzed by R. P. Poplavsky [1]. Using the canonical theory, Brillouin was able to solve a subtle physical problem concerning the functioning of the demon of Maxwell. The canonical information theory is thus an inseparable part of physics. Nevertheless, this theory does not cover re­ ception, memorizing, and generation of information. In fact, these are the processes that are essential for biology. In his posthumous monograph [21, I. I. Shmalgauzen for the first time attempted to convert Darwin’s work into a theory in terms of the canonical information theory. He introduced the feedforward and feedback channels through which the genetic and phenotypic information is transferred, and discussed the rules of encoding and transforming the biological information. This new language of the theory of evolution is substantive and has a pragmatic significance in that it further develops and clarifies its basic concepts. Suffice it to quote the following passage: “The entire mecha­ nism of natural selection can be presented in terms of infor­ mation theory as the transformation of feedback information, which is transmitted phenotypically at the level of the orga­ nization of individuals as complete systems, to the heredi­ tary information which is transmitted at the molecular level of organization through chromosomes”. Shmalgauzen also pioneered in this work the argument that “the current information theory has no techniques avail­ able to it for evaluating the quality of information, although this factor is often of decisive importance in biology. When an organism receives information from the environment, first of all it evaluates this information from the stand­ point of its quality...”. This statement is irrefutable. The quality, meaning, con-

86

M. V. Volkenshtein

tent, or value of information have indeed become the object of study in biophysics. Let us accept the term “value of information”. Obviously, this concept can be defined only in connection with the re­ ception of information because a measure of this value is given by the consequences of the reception of this informa­ tion by the organism. Hence, an analysis of the value of infor­ mation must begin with an analysis of reception. The reception of information, and hence, its storage, is a process which is in principle irreversible; it is realized when the initial state of the receptor is unstable and the receptor switches to a new, relatively stable state. The de­ finition of the quality of information involves a concept of the level of reception. This concept is related, among other factors, to the amount of information stored earlier (the thesaurus of the receptor). Reception signifies that information has been irreversibly memorized. Information can be lost (forgotten), but it can­ not be channeled back. As a result of the extremely nonequilibrium nature of the process of reception, triggered phenomena, similar to phase transitions, take place; such processes are especially important in biology. The processes are such that very small amounts of information cause substantial, macroscopic events. For instance, one bit of information carried by a change of traffic lights from red to green triggers changes in traffic flow. In the case of reception and storing the incoming informa­ tion, the equivalence of information to entropy is far from obvious. At the same time, these processes are evidently real physical phenomena calling for further analysis. In my opinion, the difficulties encountered in this field by thermo­ dynamics are of principal nature, and follow from the dif­ ficulties in dealing with irreversible processes involving

Information Theory and Evolution

87

long-term memory, i.e., processes with very long (practically infinite) relaxation time (relaxation time is the time ne­ cessary to reach equilibrium). What are entropy and thermo­ dynamical probability in these conditions? Alas, the phys­ ics of such irreversible processes has not been developed yet, despite some substantive efforts. This is also true for another phenomenon which is of ut­ most importance for biology, the generation of new infor­ mation. The creation of new information is the act of memo­ rizing the outcome of random selection. Phenomena of this type are abundant in evolution. Sexual reproduction is equi­ valent to storing the outcome of a random event, namely, the formation of a new genotype as a result of recombination of the parent genomes. This event is indeed random (and hence, somewhat free) because no law dictates that the offspring should be born to this particular pair of individuals. Incidentally, the generation of new information by crea­ tive activities, such as writing poetry, also proceeds in the manner of making a random choice (i.e. free choice) to memory. Here again we encounter an irreversible process which is difficult to interpret in terms of thermodynamics. Information thus has two, and only two, aspects that di­ rectly concern physics. The first aspect is the amount of in­ formation in equilibrium; the second one is the value of in­ formation, directly related to the process of reception and memorizing. No physical theory of these processes has been developed so far. It is clear, nevertheless, that this is a job for physics, and solely physics. Regardless of the future theory, we can accept, with re­ servations, a conditional definition of the value of informa­ tion as the degree of its non-redundancy or independence [31. Redundant or repeated information is of no value for the receptor. The protagonists of Jules Verne’s C a p t a i n G r a n t ' s

88

M. V. Volkenshtein

C h i l d r e n successfully reconstruct almost the entire text of the message extracted from a bottle, even though quite a few letters are washed off. Hence, these letters were redun­ dant; in contrast, non-redundant, irreplaceable letters have high value. With this definition of the value of information, we come to a conclusion that in the course of biological development (ontogenesis and phylogenesis) the value of information, and hence, its irreplaceability, increase. The former factor is evidenced by the transformation of presumptive rudiments into determinative ones in the course of embryogenesis, and also by the phenomenon of recapitulation. The latter factor is found in the event in which new species arise from com­ mon ancestors as a result of biological divergence. Numerous examples could be added to these two. The essential feature is that the value of information changes in the manner of phase transition. If evolution enhances the value of information, there are reasons to believe that a similar effect takes place at the molecular level. A conditional scale of the values of amino acid residues in proteins can be composed in terms of re­ sidue replaceability. Using this scale, it is possible to show that the total value of amino acid residues in cytochrome c was indeed increasing through the evolutionary tree both for mammals and for birds. This means that mutationcaused replacements substitute less valuable residues for more valuable ones. In fact, no such relationship was found in the case of hemoglobins: evolution resulted in random substitutions. These results agree with Kimura’s neutral evolution theory which holds that at the molecular level the evolution mostly proceeds in a neutral, random manner. Numerous replace­ ments of nucleotides in DNA or of amino acid residues in proteins do not feel the pressure of natural selection which

Information Theory and Evolution

89

acts at the phenotype level. Kimura’s theory is well support­ ed by evidence. Its physical meaning reduces to a non-singlevalued, degenerate correspondence of the primary protein structure, which is genetically encoded, and the spatial structure which is responsible for the biological function of the protein. In spite of the statements of some scientists, the neutralism theory in no way stands in opposition to Darwin’s theory. Nevertheless, different proteins hehave quite dissimilar­ ly. A more ancient and universal protein, the cytochrome c, is less subject to mutations than a younger protein, hemo­ globin. This explains why the roughly manifested trend of increased irreplaceability of amino acid residues is observed for cytochrome c but not for hemoglobin. In this sense, the protein evolution goes “from Kimura to Darwin”, from neu­ tralism to selectionism. It was shown that the evolutionary enhancement of pro­ tein value requires that there be a stock of residues with low value, i.e., with high replaceability. The principle of evolutionary growth in the complexity of biological systems was discussed in the literature in recent years. The complexity concept needed a rigorous de­ finition. Such a definition was given by Kolmogorov: the complexity of an object is the minimal number of binary digits, encoding the information on this object, sufficient for reproduction (decoding). In other words, the complexity is the length, measured in bits of information, of the short­ est program generating the message about the object. In order to ascertain that a given sequence of digits is complex (i.e., random), it is necessary to prove Jthat no shorter program is possible for generating this sequence. This statement cannot be proved in view of Godel’s theorem, and the proof calls for a system of greater complexity. We should emphasize at this point that science invariably

90

M. V. Volkenshtein

aims at discovering the minimal program which generates (explains) the set of facts available for analysis: this is the familiar Occam’s razor. Thus, Newton’s law of gravitation explains both the fall of an apple and the motion of planets. However, Godel’s theorem does not permit a proof of the minimality in terms of logic. This is why science is impossib­ le without intuition. Mandelshtam used to say that the Schrodinger equation was the result of guess, not of inference. The most complex systems in nature are individual liv­ ing organisms, man’s organism among them. Each living being is unique, and cannot be presented by a shorter pro­ gram. In this sense, “nobody is replaceable”. This statement holds also for man’s creative output, the pieces of literature and art. However, each organism is more than just an individual. It represents a kingdom, phylum, class, order, family, genus, and species. This is a very real hierarchy, and its discovery constituted one of the greatest events in the history of scien­ ce. The complexity is obviously increasing from kingdom to species. Within each taxon, “nothing is irreplaceable”: all representatives of a given species are interchangeable, and are described by the same minimal program. Let us turn to evolution. Usually, although not always, complexity increases in the course of phylogenesis. For instance, the transition to the parasitic existence signifies simplification, not greater complexity. The concept of complexity is relative. A bull’s brain is a fantastically complex system for the biologist who needs hundreds and thousands of bits for its description, but a butcher needs at most 5 bits, since brain is merely one of about thirty edible parts of bull’s body. We have to deal with different levels of data reception, with the relativity in the value of information. We find that complexity is equivalent to irreplaceability, or to non-

Information Theory and Evolution

91

redundancy at a given level of reception. What is irreplacea­ ble, is complex. The impossibility of a further minimization of a program which generates a complex message signifies that the program is irreplaceable. Both the information value and complexity increase in the direction from kingdom to species and reach a maximum in the individual. Never­ theless, the concept of the value of information is richer than that of complexity. The complexity refers to an object as a whole, while the value is inherent in each individual ele­ ment of the object. The complexity characterizes the struc­ ture, while the value represents the function as well. Consider the evolutionary simplification, which occurs in the evolution of vertebrates too. In four families of deepsea anglerfishes (iC a u l o p h r y n i d a e , C e r a t i d a e , N e o c e r a t i d a e , L i n o p h r y n i d a e ) the relations between sexes are very peculiar. The male, which is much smaller than the female fish (a C e r a t i s h o l b o e l l i female is more than 1 m long, while a male may be as short as 15 mm!), penetrates the skin of the fe­ male, after which its jaws, eyes, and intestines undergo reduction, so that it ultimately transforms into an append­ age producing the sperm. The simplification is indeed dras­ tic, but its outcome in these specific ecological environ­ ments is the enhancement, not decrease, of the value. The principle of value enhancement is independent of natural selection. Nevertheless, its formulation deliberately emphasizes the inherent directivity, or irreversibility, of biological evolution. An increasing value entails an enhanced ability of a bio­ logical system to extract valuable information. This ability is especially well developed in higher animals whose sensory organs specialize in information selection. The frog responds only to moving insects, the bat operates its sonar and res­ ponds only to reflected signals, not to direct ones. The se­ lection of valuable information forms the foundation of

92

M. V. Volkenshtein

man’s creative activities. This selection requires no addi­ tional energy consumption, and one bit of information is obtained at a cost independent of its value. Natural selection means that phenotypes are subjected to comparative evaluation in terms of a given ecological niche, i.e., it is a search for optimal value. The situation can be elucidated by an analogy to chess. At the initial position a chess player selects from among 20 possible moves. In reality no one with even a minimum skill searches among all these moves (“all mutations”) but analyzes at most five or six possibilities. The number of possible moves increases with each move, but the choice nar­ rows down still more. Each move creates a new “ecological environment” on the chess board. The role of mutation is played by the opponent’s move. Once a game of chess was di­ rected into a certain path, it cannot be drastically redirect­ ed. The game is irreversible; the moves cannot be retracted. Here lies the analogy to evolution, which is invariably a one­ way, or channelled process. Terrestrial vertebrates have four limbs because their ancestors, Late Devonian C r o s s o p t e r y g i i bony fishes, had four corresponding fins. Chess suggests another, more interesting analogy. Accord­ ing to Steinitz’s theory, the game must follow the positional strategy, striving to accumulate small advantages. Once the advantages grew sufficiently large, the player was to seek a combinational, resolute path to victory. The nontri­ viality of this theory, which was supported by Lasker’s detailed logical arguments, lies in the following feature: if the positional gains are not used for advantage, at the right moment, they vanish. Lasker wrote: “The combinational and positional strategies of a chess master are complemen­ tary. A combination serves to invalidate false values, and the positional game is aimed at solidifying and utilizing the true values” [41.

Information Theory and Evolution

93

Lasker regarded chess as a model of “life’s struggle”. He did not notice that chess offered a model of natural se­ lection and struggle for survival: the accumulation of small advantages is similar to microevolution, and the transition to a combination is like macroevolution, or phase transi­ tion. One of the basic problems in evolution theory is the reducibility of macroevolution to microevolution. The above arguments are very similar to the concepts of idioadaptations and aromorphoses introduced by Severtsov [5]. The terminology used by chess players is in itself an in­ dication of the analogy. Botvinnik wrote: “Euwe was able to adapt himself to a situation arising in a game... An ana­ lysis of ‘adaptive evaluations’ in a game of chess will lead to the rise of the perfect chess robot” [61. The directivity of evolution is imposed by both exogeneous and endogeneous factors. The exogeneous factors are the ecosystem and natural selection. The endogeneous factors stem from the program of ontogenetic development incor­ porated in each organism; this program imposes rigid bounds on the evolutionary diversity. The problems of evolution are inseparable from those of ontogenesis, and the interrela­ tion of ontogenesis and phylogenesis can, in fact must, be considered in the framework of information value concepts. The notion of the value, or irreplaceability of information is a fundamental one. Note that the “technology” of biological evolution differs from the man-made technology. We tend to construct machines from mutually replaceable parts, com­ posing sets of spare parts. For any organism more perfect than a hydra or sponge, the decisive factor is irreplaceabil­ ity at all levels, up to the immunological level in higher vertebrates. The rejection of “block technology” and “spare parts”, giving up the capability of regeneration, is the price for the sophisticated holistic organization, for the welldeveloped nervous system.

94

Yu. M. Svirezhev

References 1. IIonjiaBCKHu P. II. TepMOflHHaMHKa HHopMai*iioHHbix npoueccoB. M.: Hayna, 1981. 2. IIlMajibray3eH H. H. Kn6epHeTHHecKne Bonpocu 6nojioriin. Hobo-

ch6hpck: HayKa, 1968. 3. B ojibK eH iuT eH H M. B. Bno(J)H3HKa. M.: HayKa, 1981. 4. Jlacnep 3. yneCuuK m axM aT H O H nrpu. M.: Ou3nyAbmypa u m y pu3M, 1937. 5. CeBepuoB A. H. rjiaBHwe HanpaBJieHHH aBOJiKmuoHHoro nponecca. M.: M3fl-B0 M ry, 1967. 6. Botbuhhhk M. M. Ot iiiaxMaTHCTa k MaiiiHHe. M.: &u3Kyjibmypa u cnopm, 1979.

Control Sciences and the Harvest YU. M. SVIREZHEV

This article is an attempt to approach the problem from different angles, e.g., that of engineers and ecologists, and pose some questions which can hardly be answered in un­ ambiguous terms. Until recently the agriculturists did not need mathemat­ ics other than ordinary statistics and design of experiment. It is curious, however, buj, it were the needs of agriculture that gave an impetus to exploring these mathematical fields. Thus Ronald Fisher, one of the founders of today’s statistics and experimental design theory, had closely cooperated for years with selectionists at Rothamstead experimental sta­ tion in Britain. Control engineering, optimization principles, and comput­ er control were originally applied in manufacturing indu­ stries and greatly enhanced the productivity of labor. Agri­ culture was initially left out, probably for two reasons: (1)

Control Sciences and the Harvest

95

unlike industrial processes, biological systems cannot be easily described in formal mathematical terms, and (2) owing to its ancient origin agriculture is more conservative and so tends to preserve its centuries-old practices. But the growing needs called for dramatic changes. This was the case when mineral fertilizers were intro­ duced in the middle of the 19th century. At that time J. Liebig approached the stepping up of agricultural pro­ duction as a problem in engineering, and formulated a typ­ ical control-science-like principle of limiting, factors. Known today as the “bottleneck” (critical path), this principle is widely applied to process control and economy management. Agriculture Viewed by Engineering. Harvest Programming An engineer may regard agriculture (in the narrow sense of the word, i.e., crop growing) as obtaining a product with the use of some mineral* raw materials and solar energy with a certain efficiency. The principal technology is photosynthe­ sis. The process is controlled by varying the input flow of the raw material; the output, the harvest, is maximized in actual environmental conditions by optimizing control. This, roughly, is a definition of the “harvest programming” approach. A process cannot be controlled, however, unless its model is available. In our case a green plant or a vegetation po­ pulation has to be modeled. Let us consider Rachko’s model. The growth of a hypothetical plant is described as the dynamics of the biomasses of its leaves, stalk, and roots. The exogenous variables (factors) are the photosynthetically active radiation (PAR), the air temperature, the water sup­ ply in the environment, the carbon dioxide content in the ambient air, and the concentration of mineral elements

96

Yu. M. Svirezhev

(nitrogen, phosphorus, calcium, and sodium) in the soil. The internal variables are the biomasses of leaves, stalk, and roots and the concentration of nitrogen, phosphorus, cal­ cium, and sodium inside the plant. In a very crude form the conceptual diagram of this model is shown in Fig. 13. The flows of substance and energy are represented as solid lines, thick and thin, respectively, and control data transmission links, as dashed lines. When carbon dioxide and nutrients are available and the temperature is right, the PAR-initiated photosynthesis

Fig. 13. Conceptual diagram of a plant model Pi = p4 are the C 02 concentration, PAR, water potential, and temperature, respectively; P is a vector with components p, = p4; xx = x3 are the biomasses of leaves, stalk, and roots, respectively; X is a vector with components xx = ay, 1, 2—energy and substance flows; .?—data flow; 4—flow regulator; 5—respiration

Control Sciences and the Harvest

97

produces new organic substances which are distributed in all organs of the plant. Simultaneously the energy for sustain­ ing the life processes in the plant (such as the transport of the substances, biochemical processes, etc.) is released through the oxidation of these substances in respiration. The rate of these processes is dictated by the biomasses and geometry of the plant organs and by the exogenous varia­ bles. Besides, the plant has its own control mechanisms such as stomatal transpiration and distribution of the assimilants. Many of these mechanisms remain little explored and so the model loop is closed by using either empirical func­ tions or general biological considerations which are formulat­ ed as some maximum principle such as adaptation. On the other hand, the dependence of photosynthesis on PAR, on availability of water, and on the concentration of nutrients has been thoroughly explored and can be readily integrated into the model. Many variables and functions in the model have specific values only for specific plants. This is especially impor­ tant in determining the geometric characteristics of an in­ dividual plant or a population. Thus the PAR distribution is dependent on the sowing geometry, i.e., a population variable, more than on the plant geometry. The choice of specific values is the stage of model identi­ fication at which it is adjusted for controlling a specific crop. At subsequent stages the optimal control of the agricultural system is sought by varying the dynamics of watering, ferti­ lization, etc. under specified environmental conditions beyond human control, such as ambient temperature and rainfall. This approach has been used in the Computing Center of the USSR Academy of Sciences by R. Saidulloev and A. M. Tarko who developed a cotton growth model. Thus, numerous computer experiments suggest that periodic water7-0913

98

Yu. M. Svirezhev

ing results in better harvests than continuous watering does. Now, a question may arise, why bother with a model which cannot recognize all the factors when an optimal com­ bination of control parameters can be obtained by experi­ mental design? First, every experimental run would require an inadmissibly large surface and so would be costly and take too much time. Second, in experiment the exogenous conditions could not be monitored as closely as, for instance, in petrochemical synthesis where the response follows in a matter of seconds and the entire process is observable. Simulation methods are therefore preferable. In the Netherlands models of wheat and corn growth have been experimentally tested. In the USSR harvest program­ ming is the subject of vigorous research by E. P. Galyamin, R. A. Poluektov, Yu. K. Ross, and 0. D. Sidorenko. Agricultural Viewed by an Ecologist. Monoculture or Agrocenosis?

An approach in which agriculture is viewed as a kind of manufacturing industry, the plant as a kind of machine, and the harvest control as an engineering task is legitimate at a certain stage because it results in a sharp increase of farming production. However, a price should be paid for this increase. The production cost estimated in terms of total energy and mineral inputs expressed in energy units per ton of wheat has increased 50-fold during the last 100 years. In this sense the highly industrialized US agriculture is 250 times costlier than the traditional agriculture of South-East Asia. Programmed harvesting on a larger scale would be still more expensive because a price would have

Control Sciences and the Harvest

99

to be paid for both the indispensible information on the system state and for control itself, i.e., for maintaining the system at the computed optimal level. At this point the desire to obtain maximal productivity is in conflict with the maximal stability requirement made by nature to every bilogical community, be it population or biocenosis. One ecological indicator of the community’s stability is its variety. This is essentially a measure of the amount of information describing the community. The stability in face of environmental variations increases with the number of species having different characteristics. On the other hand, the productivity of a community is maximized when all the individual characteristics are made to approach some optimal value. In this way, however, the variety is reduced. Monoculture is ideally optimal but also absolutely unstable because there is no variety in it. It is human control that keeps monoculture stable. The evolution of natural communities increases their variety. The community pays for it by increased energy dis­ sipation. Any exploitation such as collection of a harvest and removal of some biomass from the community reduces the dissipation, and, consequently, the stability. So, harvesting should be discontinued if stability is to be maximized and if it provides a maximal harvest the com­ munity is absolutely unstable. From this point of view the agrotechnology of high yield crops is but stabilization of the unstable monoculture population. A compromise should probably be sought whereby mono­ culture is replaced by an agrocenosis (agroecosystem), a man­ made community whose structure would be fairly similar to that of natural communities and whose stability would be largely maintained by inherent ecological regulatory mecha­ nisms rather than a flow of man-supplied energy. 7*

Yu. M. Svirezhev

100

Development of ecologically optimal agricultural struc­ tures is today on the agenda of mankind. Let us consider a modeled case study. Theory of Trophic Chains. Harvest versus Stability Let us start with some statistics. According to various estimates, about one third of the entire US harvest is lost Q

Fig. 14. Trophic chain of an agro­ ecosystem Q—resource arrival; 1-3—chain levels

to pests, chiefly insects. The direct pest control costs amount to about 2,000,000 dollars annually. What are the environmental consequences of this combat?

Control Sciences and the Harvest

101

Figure 14 shows the trophic chain of any crop. The new trophic level is that of pests, such as the cotton-worms; it is followed by the parasites or predators of these pests, such as various ichneumon flies. The resources are mineral ferti­ lizers Q . The harvest would seem to increase with Q as

Fig. 15. Harvest as a typical function of fertilization

shown in Fig. 15. However, larger Q do not necessarily improve the harvest. The trophic chains are found to be discrete, or resource-quantized. Indeed, there are critical values